# Sleev Source: https://sleev.ai This file contains rendered public Sleev site content intended for LLM consumption. ## Included pages - [Sleev | AI Context Optimization & Token Cost Reduction](https://sleev.ai/) - [Performance & Benchmarks: DeepSWE Token Savings | Sleev](https://sleev.ai/performance) - [Pricing: Free & Pro Plans for Coding Agents | Sleev](https://sleev.ai/pricing) - [Enterprise AI Context Management & Cost Control | Sleev](https://sleev.ai/enterprise) - [Data Privacy | Sleev](https://sleev.ai/data-privacy) - [Enterprise Pitch Deck | Sleev](https://sleev.ai/deck) - [Quickstart | Sleev Docs](https://sleev.ai/docs/quickstart) - [What is Sleev | Sleev Docs](https://sleev.ai/docs/what-is-sleev) - [Install - Quickstart | Sleev Docs](https://sleev.ai/docs/quickstart/install) - [Setup - Quickstart | Sleev Docs](https://sleev.ai/docs/quickstart/setup) - [Configuration | Sleev Docs](https://sleev.ai/docs/config) - [Prerequisites - Configuration | Sleev Docs](https://sleev.ai/docs/config/prerequisites) - [Config Wizard - Configuration | Sleev Docs](https://sleev.ai/docs/config/config-wizard) - [Providers - Configuration | Sleev Docs](https://sleev.ai/docs/config/providers) - [Custom providers - Configuration | Sleev Docs](https://sleev.ai/docs/config/custom-providers) - [URL-based routing - Configuration | Sleev Docs](https://sleev.ai/docs/config/url-routing) - [Gateway bind address - Configuration | Sleev Docs](https://sleev.ai/docs/config/gateway-bind) - [Limited internet - Configuration | Sleev Docs](https://sleev.ai/docs/config/limited-internet) - [Deconfigure - Configuration | Sleev Docs](https://sleev.ai/docs/config/deconfigure) - [Harness Setup | Sleev Docs](https://sleev.ai/docs/harness-setup) - [Claude Code CLI Setup & Token Cost Optimization | Sleev Docs](https://sleev.ai/docs/harness-setup/claude-code) - [Claude Desktop Setup & Token Cost Optimization | Sleev Docs](https://sleev.ai/docs/harness-setup/claude-desktop) - [Codex / ChatGPT Setup & Token Cost Optimization | Sleev Docs](https://sleev.ai/docs/harness-setup/codex) - [OpenCode Setup & Token Cost Optimization | Sleev Docs](https://sleev.ai/docs/harness-setup/opencode) - [Pi Setup & Token Cost Optimization | Sleev Docs](https://sleev.ai/docs/harness-setup/pi) - [Grok Build Setup & Token Cost Optimization | Sleev Docs](https://sleev.ai/docs/harness-setup/grok-build) - [Hermes Agent Setup & Token Cost Optimization | Sleev Docs](https://sleev.ai/docs/harness-setup/hermes-agent) - [Premium Requests | Sleev Docs](https://sleev.ai/docs/premium-requests) - [Premium Models | Sleev Docs](https://sleev.ai/docs/premium-models) - [Offline Licensing | Sleev Docs](https://sleev.ai/docs/offline-licensing) - [When to use it - Offline Licensing | Sleev Docs](https://sleev.ai/docs/offline-licensing/when-to-use) - [Create a license - Offline Licensing | Sleev Docs](https://sleev.ai/docs/offline-licensing/create) - [Renew - Offline Licensing | Sleev Docs](https://sleev.ai/docs/offline-licensing/renew) - [Limits and recovery - Offline Licensing | Sleev Docs](https://sleev.ai/docs/offline-licensing/limits) - [Privacy - Offline Licensing | Sleev Docs](https://sleev.ai/docs/offline-licensing/privacy) - [CLI Reference | Sleev Docs](https://sleev.ai/docs/cli-reference) - [FAQ | Sleev Docs](https://sleev.ai/docs/faq) - [Privacy Policy | Sleev](https://sleev.ai/privacy) - [Terms of Service | Sleev](https://sleev.ai/terms) - [EULA | Sleev](https://sleev.ai/eula) - [Data Processing Addendum (DPA) | Sleev](https://sleev.ai/dpa) - [Subprocessors Directory | Sleev](https://sleev.ai/subprocessors) - [Security | Sleev](https://sleev.ai/security) - [Contact Us | Sleev](https://sleev.ai/contact) --- ## Sleev | AI Context Optimization & Token Cost Reduction URL: https://sleev.ai/ Description: Token management for the agentic era. Sleev's AI-assisted context optimization reduces token spend and maximizes agent session value. tokens saved # Sleev, fix your AI bill cheaper requests, smarter context, same model Bash PowerShell Npm Agent $ curl -fsSL https://sleev.ai/install.sh | bash Local gateway • No prompt collection Works with Claude Code OpenCode Pi Grok Build Hermes Agent Codex ChatGPT GitHub Copilot Kimi Z.AI Claude Code OpenCode Pi Grok Build Hermes Agent Codex ChatGPT GitHub Copilot Kimi Z.AI Why Sleev ## A local context optimization layer for your agentic stack Sleev optimizes context so your coding agents can do more for less. It runs locally, keeping your session data private and costs down as sessions grow. full session history full session history optimized context optimized context ### Stop runaway costs Every turn re-sends the entire session history. As sessions grow, most of what you pay for is context the model has already absorbed. ### Background optimizations Sleev optimizes session context when the expected savings outweigh provider-cache update costs. ### Keep your tools and providers Same coding tool, provider, and API keys. Connect through Sleev once and keep working. Session 0 tokens Saved 0 ## Ready to fix your AI bill with Sleev? Install Sleev and start saving on your next session. [Create an account](/sign-up) or [Join Discord](https://discord.gg/W36rxJswbG) --- ## Performance & Benchmarks: DeepSWE Token Savings | Sleev URL: https://sleev.ai/performance Description: Measured Sleev results: DeepSWE 1.1 aggregates and real session curves. A quarter fewer input tokens, a lower bill, solve rate held. # Measuring Sleev It's easy to assume big agent bills are just the cost of doing serious work. But most of what you're paying for isn't new thinking. It's old terminal logs and stale error messages being re-sent over and over. Sleev cleans up that clutter automatically as you work, letting you run longer sessions for a fraction of the cost. Below, we break down our head-to-head benchmarks to show exactly how much you save, and why your agent's coding accuracy stays just as sharp. ## DeepSWE We measure Sleev on [DeepSWE 1.1](https://deepswe.datacurve.ai/), a benchmark of 113 real-repository coding tasks scored by the benchmark's own tests. We run the baseline DeepSWE and compare it to runs using Sleev as well as other popular context and cost optimization tools. We first ran the standard DeepSWE 1.1 as published: every task gets a fresh context window, so sessions stay short. However, real agentic use may often run long, and context may build up over hours of work. To measure that, we also ran a modified version that batches six tasks into one continuous context window. While solve rate holds in both and Sleev imposes no penalty on short sessions, we show that its savings grow quadratically with session length. Standard runs [GPT-5.5 high] one task per session Batched runs [GPT-5.6 medium] six tasks per session Baseline Sleev Baseline Sleev Solve rate pass@1 57.5% ± 5.0 59.1% ± 2.2 61.1% ± 2.3 61.9% ± 1.9 Cost per task mean $3.71 $3.50 $18.08 $3.99 Total cost mean per full run, 113 tasks $419.10 $395.09 $2,042.63 $451.06 Input tokens mean per task 4,380,061 3,253,863 18,566,080 3,491,998 Peak context size median per task · mean per run 102,315 82,225 674,822 141,739 Agent steps mean per task 52.5 57.0 67.7 60.2 Cache hit rate 96.0% 92.5% 99.2% 93.9% Output tokens mean per task 24,271 25,501 24,911 26,672 Standard runs: one task per context window running GPT-5.5 at high reasoning. Batched runs: six tasks chained through one continuous context window running GPT-5.6 Sol at medium reasoning. Both runs were run 4 times each. The standard runs were completed before the GPT-5.6 family was released. ## Cumulative session cost and context growth Every request an agent makes re-sends the whole conversation. In a short session that is cheap. But every turn makes the conversation longer, so every request gets more expensive than the last, and the session's total cost grows quadratically with its length. Our batched runs show this directly: with six tasks in one context window, the baseline's bill curves upward turn after turn. With Sleev the same session grows linearly. The chart also compares three other context tools run through the same batched benchmark: [Headroom](https://github.com/headroomlabs-ai/headroom), a compression layer that advertises 15 to 20% fewer tokens for coding agents; [Context Mode](https://github.com/mksglu/context-mode), which sandboxes tool output and advertises up to 98% smaller tool results; and [RTK](https://github.com/rtk-ai/rtk), a CLI proxy that filters command output and advertises 60 to 90% fewer terminal tokens. While RTK improved session cost ($95.47) and speed over the baseline, both Headroom and Context Mode ended more expensive. Sleev and Headroom both matched the baseline's 61% solve rate. RTK performance fell to 58%, while Context Mode solved 54%. Session cost GPT-5.6 Sol Averaged across full batched DeepSWE 1.1 benchmark runs: 4 runs (452 total tasks) each for Sleev and the baseline, and 1 run (113 tasks) each for Headroom, Context Mode, and RTK, chained 6 tasks per session. In multi-step agent workflows, context bloat causes more than just quadratic billing: it slows down generation, inflates time-to-first-token, and causes retrieval degradation as hundreds of stale tool outputs dilute the model's attention. In the batched baseline DeepSWE runs, input context without management generally climbs throughout the session, ending at an average of 462k tokens on turn 398. With Sleev continuously optimizing stale execution logs into high-signal context, average context stays below 100k tokens at every literal turn and ends at 74k on turn 338, 83.9% lower than the baseline endpoint. By contrast, Headroom reaches 516k tokens on turn 367. Context Mode ends at 570k tokens on turn 364. RTK reaches 565k tokens on turn 300. Note that the newest GPT models maintain unusually stable retrieval and reasoning at large context sizes compared to most architectures. While we do not yet have benchmark evidence showing solve-rate gains from context reduction on other model families, performance degradation at high context is well documented[[1](#ref-1)–[4](#ref-4)]. Session context GPT-5.6 Sol ## What about cache? Prompt caching offers substantial discounts on repeated prefix tokens, creating the intuition that retaining bloated context in the prompt is cheap. In long multi-step agent sessions, however, cache reads compound quickly. In baseline benchmarks, cache read fees account for 88% of total inference spend ($95 out of $107 per session) as the window expands. Sleev compressions break the prompt cache, requiring a one-time rewrite (an uncached write fee on that turn). But they shrink the prefix permanently, lowering the cost floor of every subsequent request. The upfront rewrite cost breaks even within 2 to 3 turns in this benchmark, though this depends on model pricing and performance. After breakeven, every subsequent request generates compounding net savings. Additionally, Sleev keeps per-request pricing predictable, avoiding the 7x higher cost volatility seen in the baseline as unmanaged context expands. Cost per request Single-turn cache rewrite fee vs. ongoing per-turn savings The baseline cost jump at request 174 is attributed to OpenAI API rate increases when prompt context exceeds 272k tokens. ## Methodology All benchmarks are measured directly against real-repository coding tasks using identical seeds, pinned harness versions, and sandboxed execution environments. The full evaluation harness and replication tooling are available in the [benchmark repository](https://github.com/sleev-labs/benchmark). Unlike Sleev, which maintains full feature parity across every supported coding harness, competing tools vary in their capabilities and stability depending on the environment. To ensure a fair evaluation, we selected the best-supported harness configuration as documented by each competing tool. ## References - [Chroma Research (Context Rot)](https://www.trychroma.com/research/context-rot) : Empirical report evaluating retrieval degradation across frontier models as context windows fill. - [Stanford "Lost in the Middle" (arXiv:2307.03172)](https://arxiv.org/abs/2307.03172) : Academic foundation establishing position-based performance degradation in long contexts. - [NVIDIA "RULER" (arXiv:2404.06654)](https://arxiv.org/abs/2404.06654) : Benchmark evaluation demonstrating that effective context windows fall far short of advertised maximums. - [GAIR "Context Rot in Long-Horizon Search" (arXiv:2606.29718)](https://arxiv.org/abs/2606.29718) : Multi-turn agentic study evaluating degradation over extended horizons. ## Footnotes - Session-context curves summarize committed GPT-5.6 archives (76 unmanaged, 76 with Sleev, and 19 each with Headroom, Context Mode, and RTK). The independent-task DeepSWE results are a separate experiment run on GPT-5.5. - DeepSWE figures are not comparable to official leaderboard scores: Codex CLI replaces DeepSWE's reference harness. Runs from June 21 to July 2, 2026. - 95% confidence intervals follow DeepSWE convention: 1.96 times the standard deviation of run-level pass@1 over the square root of runs. - Batched-run costs are repriced per request at OpenAI's list (non-promotional) GPT-5.6 Sol rates: $5 / $0.50 / $30 per million input, cached input and output tokens, doubling input and adding 50% to output for any request above 272K input tokens. Sleev's costs include the gateway's own compression calls. --- ## Pricing: Free & Pro Plans for Coding Agents | Sleev URL: https://sleev.ai/pricing Description: Choose a Sleev plan based on monthly premium requests. Most models are free, while a hand-picked set of models is premium. # Pricing plans Monthly Annual 2 months free For side projects ## Indie $ 0 per month [Install](/docs/quickstart/install) No credit card needed - 2,000 premium requests - Unlimited free requests - Personal dashboard - Community support For frequent use ## Starter $ 5 per month [Get Starter](/checkout?tier=starter&interval=monthly) Billed monthly Try again - 10,000 premium requests - Unlimited free requests - Personal dashboard - Community support For power users ## Pro $ 20 per month [Get Pro](/checkout?tier=pro&interval=monthly) Billed monthly Try again - Unlimited requests - Early access to features - Personal dashboard - Priority support For organizations ## Enterprise Talk to our team [Contact sales](/contact) - Organization management - Centralized usage reporting - Priority compliance support - Offline licenses ## Requests SLEEV PRICING MODEL ## What are Sleev's request types? Requests are classified as free or premium. Most models are free, while only a hand-picked set of models is premium. ## What counts toward my quota? A request counts toward your quota only when it uses a Premium model and Sleev saved tokens on that request. Requests to free models, and requests where Sleev saved nothing, never count. ## Can a model's request type change? Yes. We may update the list of free and premium models over time. ## Where can I track my quota? Track your premium request usage in your dashboard or the Sleev terminal application. ## What do I share with Sleev? Sleev collects account and operational information, including usage and savings metrics. We do not collect your code, prompts, or provider API keys. [Read our Privacy Policy.](/privacy) Model Request type Loading models… Enable JavaScript to load the current model catalog. Try again Previous Page 1 Next --- ## Enterprise AI Context Management & Cost Control | Sleev URL: https://sleev.ai/enterprise Description: Local context optimization for AI coding teams. Keep your tools and providers, and scope a Sleev pilot around your own workloads. # Your agentsLess overhead Local context optimization for AI coding teams. Keep your tools and providers. Evaluate the difference on your own workloads. Discuss a pilot ## Start with your workload Try Sleev with your team, using your existing setup. See how it affects costs and day-to-day coding. [Explore the benchmarks](/performance) Team use requires an enterprise agreement. ## Clear data boundaries Context processing runs on your infrastructure. Sleev receives operational and meta data. We do not use customer data to train AI models or sell it. DPA applies when agreed in writing. Restricted network? [Explore offline licensing.](/docs/offline-licensing/when-to-use) --- ## Data Privacy | Sleev URL: https://sleev.ai/data-privacy Description: Local context optimization. No prompt collection, no AI training, no data sales. Learn what Sleev collects to operate and improve the service. # Your prompts.Your data. Sleev optimizes context locally on your device. Your coding sessions do not pass through Sleev's servers. Local processing ## Your work stays with you. We do not collect your prompts, code, conversation history, model responses, tool content, or provider API keys. Sleev runs as a local daemon on your machine. Context optimization, caching, and session history stay strictly on your device, communicating directly with your upstream providers. What we collect ## Information to run and improve Sleev. Account & billing To sign you in, manage your plan, and process payments. Usage & performance Session and request identifiers, performance metrics to measure savings, administer usage limits, and improve features and reliability. Support & diagnostics Messages and optional operational logs for troubleshooting. Diagnostic uploads exclude coding content and provider API keys. ## No AI training.No data sales. We do not use customer data to train AI models or sell personal data. You maintain control of your data. You can delete your account from your dashboard settings or contact us anytime to request deletion or object to processing. --- ## Enterprise Pitch Deck | Sleev URL: https://sleev.ai/deck Description: Sleev is the context management layer for enterprise AI coding. Cut wasted LLM spend while preserving long-session task success. # Token management For the agentic era Cut your LLM bills, increase long-horizon task success and session value. Keep your harnesses, providers and keys. Built for teams turning agents into durable engineering leverage. Unmanaged context With Sleev Sleev's optimization pipeline runs locally in the background Scroll ←→↑↓ 01What Is Sleev ## A local context optimization layer for your agentic stack A local context optimization layer for your agentic stack Sleev's gateway sits between your harnesses and providers, managing and optimizing session context in a cache-efficient manner. Optimizations run in the background and don't disrupt your workflows. The result: cheaper sessions that last longer while remaining in the smart zone, where attention naturally maps to signal. Harness Claude Code OpenCode Pi Grok Build Hermes Agent Codex Sleev gateway Runs on your infra Unmanaged context With Sleev Provider Anthropic OpenAI xAI OpenRouter Read our [privacy practices](/privacy). Same harnesses Same providers Same keys 02Cost problem ## The most valuable agent sessions are also the most expensive As agents get more capable, they also get more expensive to run. Bigger context windows and longer sessions inherently turn progress into compounding token spend. Sleev is on a mission to fix this. - Unmanaged agentic work is margin loss. - The overage hides inside normal agent work. Regular agentic session Unmanaged session With Sleev Every request sends full history, compounding costs T01 T02 T03 T04 T05 T06 T07 New request Re-sent history 03Quality problem ## Smaller context is not better context Less only helps when the right context survives. Pruning, compaction, and heuristic-based compression can remove the wrong things: constraints, decisions, tool results... A good system has to distinguish disposable material from load-bearing one before it cuts. Naive Compaction Compaction is one-shot, heavily lossy, and disrupts the flow. Heuristic-based Pruning Compression Some methods trim the oldest, largest, or everything first, and can be blind to what context the task still depends on. Sleev AI-assisted Adaptive compression Sleev's background optimization targets low-signal material while preserving task-critical data. Anatomy of a long session with Sleev Context window Recent edits 24K Current goals 22K / 20K Constraints 26K / 26K Relevant errors 20K / 25K Tool outputs 17K / 113K Stale explorations 7K / 88K Logs, retries, debugging 5K / 47K Preserved Compressed Context size -73% Unmanaged context 440K tokens With Sleev 121K tokens 04Session value ## Long sessions become durable The longer the session, the more Sleev matters. It slows cost compounding while preserving the context agents need to stay productive. Sleev turns long sessions from a liability into leverage: sharper context, better accumulated learning, and more durable agent work. Cost growth flattens Context quality holds Session value compounds Session value over time With Sleev Unmanaged session 05Long-lived sessions ## Sessions can last for days Complex work does not fit neatly into one afternoon. By continuously monitoring the context window for optimizations, Sleev changes the shape of a session. A single thread can survive multi-day work without starting over. With Sleev Without Sleev 06Pilot ## Validate Sleev in your environment Define a window, compare against your cost baseline, and survey users and workflows. 01 Pilot 5–20 Engineers 2–4 weeks Same harnesses Same keys 02 Measure Savings Token and cost delta Adoption Real usage beyond first impressions Confidence Enough clean data to make a decision 03 Decide Roll out Savings and adoption are strong enough to expand. Tune Savings are real, usage needs work Stop Pilot does not earn its place ## Sleev [Privacy](/privacy) · [Terms](/terms) · [Docs](/docs/quickstart) [Start saving](/enterprise) --- ## Quickstart | Sleev Docs URL: https://sleev.ai/docs/quickstart Description: Install Sleev and start routing your coding agent through the local proxy in minutes. # Quickstart Sleev is a context-management gateway for AI coding agents, bundled with a small command-line tool, sleev, that installs and manages it. Sleev runs locally between your coding tool and model provider, optimizing session context. Coding tools resend their entire conversation on every turn, which burns tokens quickly as a session grows. Sleev optimizes session context to reduce token usage, weighing the expected savings against provider-cache costs when deciding when to optimize. ## Usage and accounts You can use Sleev without creating an account or signing in. Running in account-free mode provides 500 free Premium requests per month with full local optimization. Signing in with a free account gives you 2,000 free Premium requests per month, and [paid plans](/pricing) offer even more. Requests that do not use [Premium models](/docs/premium-models) are always free and unlimited, with full optimization. To learn more, see our [Premium requests](/docs/premium-requests) guide. --- ## What is Sleev | Sleev Docs URL: https://sleev.ai/docs/what-is-sleev Description: Learn how Sleev optimizes context windows, cuts LLM token costs by 30-80%, and boosts agent reliability. # What is Sleev Sleev cuts your token spend by 30% to 80% per session without changing your agentic stack. Our edge: keep the signal, cut the noise, spend fewer tokens getting the same or better work done. ## The problem Agent sessions are inherently messy. Requests pile up: tool responses, skills, explorations, error backtracks, logs, file reads... The important bits get buried under noise, things get lost in the middle. As the session grows, this compounds. Requests get bigger, more expensive, and inevitably worse. At scale, this is incredibly inefficient and every request adds to this debt. ## How Sleev works By exposing a purpose-built context management toolkit to your agents, and transparently managing transient session state, your sessions get optimized with each passing request. - Deep Optimization: Tool responses, intermediate steps, inactive explorations, redundant file reads, bash commands... Every session part is intelligently optimized, at the right time. - Guidance: Your agents just know what to do, and how best to leverage Sleev. - Non-Intrusive: By operating in the background, Sleev abstracts all optimization away for a seamless experience. ## Prompt caching and cost Prompt caching discounts repeated tokens, but unmanaged sessions still spend the bulk of their budget repeatedly reading expanding context. Sleev optimizes session context to reduce ongoing inference costs. For measured benchmarks, cache economics, and turn-by-turn cost breakdowns, see our [Performance report](/performance) . ## Smarter sessions Smaller context is not automatically better. Simple compaction just loses information. Sleev is built around keeping recall useful. By filtering out dead-ends, repetitive tool outputs, and stale context, it reduces distractions. The model can focus on the current goal and immediate constraints. That leads to fewer hallucinations, fewer runaway loops, and higher task completion rates. --- ## Install - Quickstart | Sleev Docs URL: https://sleev.ai/docs/quickstart/install Description: Install the Sleev command-line tool on macOS, Linux, or Windows. # Install Start by installing the sleev command-line tool, which you use to set up and manage the gateway. The native installer downloads the correct binary, verifies it, and installs it in your user account without requiring administrator access. If the install directory is not on your PATH, it prints the command needed to add it. Sleev is proprietary software. Product use is governed by the [Terms](/terms), the [EULA](/eula), and the [Privacy Policy](/privacy). ## macOS, Linux, and WSL ~~~ curl -fsSL https://sleev.ai/install.sh | bash ~~~ ## Windows PowerShell ~~~ irm https://sleev.ai/install.ps1 | iex ~~~ If you prefer npm, you can install sleev with npm instead, which works on macOS, Linux, and Windows: ~~~ npm install -g sleev ~~~ --- ## Setup - Quickstart | Sleev Docs URL: https://sleev.ai/docs/quickstart/setup Description: Sign in, install the local Sleev gateway, and keep it up to date. # Setup With sleev installed, run it to open the TUI. ~~~ sleev ~~~ ## Sign in The first time you run sleev , it asks you to sign in to your account. This links the gateway on your machine to your Sleev account so your usage and metrics show up in your dashboard. Signing in is optional. You can also run Sleev in account-free mode with 500 free Premium requests per month. Hello John, let's start saving tokens Sign in join Discord Sleev has saved 35 402 023 193 tokens worldwide ### Command-line setup If you want to use only the CLI, run each step from the terminal. Sign in to your Sleev account: ~~~ sleev auth login ~~~ Register this machine, install the gateway if needed, and start it: ~~~ sleev setup ~~~ ## Check the gateway After setup, confirm the gateway is running locally: ~~~ sleev gateway status ~~~ CLI and gateway release archives include LICENSE.md, EULA.md, and THIRD_PARTY_NOTICES.md next to the binaries. ## Stay up to date Each time you open sleev , it checks for a newer release. Sleev does not update automatically. If a release is available, a notification appears in the corner with the new version and an Update button that upgrades the CLI and gateway together. ● running Sleev v1.4.6 update available - 1.4.7 Update Or update from the command line at any time: ~~~ sleev upgrade ~~~ If you want to pin a version, do so like this: ~~~ sleev upgrade 1.2.0 ~~~ --- ## Configuration | Sleev Docs URL: https://sleev.ai/docs/config Description: Set up the local Sleev gateway, providers, and shared routing options. # Configuration Configuring Sleev means running the local gateway and telling your harnesses which provider traffic should use. Choose a topic in the navigation for detailed instructions. The Config Wizard in the sleev TUI handles the common path. --- ## Prerequisites - Configuration | Sleev Docs URL: https://sleev.ai/docs/config/prerequisites Description: What you need before configuring a Sleev harness. # Prerequisites Before you configure a harness, make sure the gateway is installed and running. The [Quickstart](/docs/quickstart) walks through this in full. You should have: - Installed the sleev command-line tool. - Run setup to register this machine, install the gateway, and start it. - Confirmed the gateway is running locally on 127.0.0.1:17321. You can check that the gateway is up at any time with: ~~~ sleev gateway status ~~~ --- ## Config Wizard - Configuration | Sleev Docs URL: https://sleev.ai/docs/config/config-wizard Description: Configure harnesses and providers with the Sleev TUI or CLI. # Config Wizard The easiest way to configure any harness is the Config Wizard built into the TUI. It applies the appropriate configuration for each supported harness. - In the TUI, open Config Wizard. It lists the harnesses detected on your machine. ● running Sleev v1.6.16 Menu (m) ← back (esc) ● ▾ Claude Code 2.1.233 ● Anthropic ● Manual setup guide OPEN DOCS › Codex 0.147.0 ● › Hermes Agent - › OpenCode 1.18.18 ● › Pi 0.82.1 ● › Grok Build 1.0.4 ● - Select the harness you want to configure. Sleev shows the providers supported for it. - Enable the provider you want to use. Sleev writes the appropriate harness setting. Use [Harness Setup](/docs/harness-setup) if you prefer to edit it yourself. ## Command-line setup Open the same harness and provider picker without launching the full TUI: ~~~ sleev harness ~~~ Scripts and coding agents can use sleev harness list, then inspect and configure one harness with sleev harness list and sleev harness enable [provider]. Claude Code and Codex select their provider automatically; other harnesses require a provider ID from their list. For the full command list, see the [CLI Reference](/docs/cli-reference). --- ## Providers - Configuration | Sleev Docs URL: https://sleev.ai/docs/config/providers Description: Providers supported by each Sleev harness. # Providers Built-in provider support depends on the harness. If you need a compatible provider that is not listed, [use a custom base URL instead](/docs/config/custom-providers). Claude Code harness ID: claude-code anthropic Anthropic Claude models, with a pay-as-you-go API key or a Claude Max subscription. Codex / ChatGPT harness ID: chatgpt openai OpenAI models with a pay-as-you-go API key. codex OpenAI's Codex backend through ChatGPT sign-in, with no separate API key. OpenCode harness ID: opencode anthropic Anthropic Claude models with a pay-as-you-go API key. Claude Max subscriptions are not supported here. openai OpenAI models with a pay-as-you-go API key. codex OpenAI's Codex backend through ChatGPT sign-in, with no separate API key. copilot GitHub Copilot models through the Copilot API. google Google Gemini models through the Gemini API. moonshotai Moonshot AI's general API for Kimi models. kimi-coding Moonshot's Kimi coding plan subscription. mimo Xiaomi MiMo Token Plan Singapore. bailian-token-plan Alibaba Cloud Bailian Token Plan. zai-coding-plan Z.AI's coding plan API for GLM coding models. opencode OpenCode Zen. opencode-go OpenCode Go. Pi harness ID: pi anthropic Anthropic Claude models. openai OpenAI models with a pay-as-you-go API key. codex OpenAI's Codex backend through ChatGPT sign-in. google Google Gemini models through the Gemini API. kimi-coding Moonshot's Kimi coding plan subscription. moonshotai Moonshot AI's general API for Kimi models. mimo Xiaomi MiMo Token Plan Singapore. bailian-token-plan Alibaba Cloud Bailian Token Plan. zai-coding-plan Z.AI's coding plan API for GLM coding models. opencode OpenCode Zen. opencode-go OpenCode Go. Grok Build harness ID: grok-build grok Grok models through your Grok subscription sign-in. xai Grok models with a pay-as-you-go xAI API key. Hermes Agent harness ID: hermes anthropic Anthropic Claude models with a pay-as-you-go API key. openai OpenAI API models with a pay-as-you-go API key. openrouter Models available through OpenRouter. google Google Gemini models through the Gemini API. kimi-coding Kimi Coding with a Kimi Code API key. mimo Xiaomi MiMo Token Plan Singapore. bailian-token-plan Alibaba Cloud Bailian Token Plan. zai-coding-plan Z.AI's coding plan API for GLM coding models. opencode OpenCode Zen. opencode-go OpenCode Go. --- ## Custom providers - Configuration | Sleev Docs URL: https://sleev.ai/docs/config/custom-providers Description: Route compatible custom provider endpoints through Sleev. # Custom providers Use custom providers when your endpoint uses an OpenAI-compatible, Anthropic-compatible, or Gemini-compatible format but is not one of Sleev's built-in providers. Most users should leave this unset. Keep the harness pointed at the local gateway, then send sleev-harness and sleev-base-url in its request headers. See the per-harness examples in [Harness Setup](/docs/harness-setup). Do not set sleev-provider for custom providers: Custom headers ~~~ sleev-harness: opencode sleev-base-url: https://provider.example.test/v1 ~~~ Use a full http or https base API URL. Include a path such as /v1 only when it is part of the provider's API base. Hermes custom upstreams use a named provider with the complete nested upstream in its API URL instead of routing headers. See the [Hermes Agent setup](/docs/harness-setup/hermes-agent) for an example. --- ## URL-based routing - Configuration | Sleev Docs URL: https://sleev.ai/docs/config/url-routing Description: Route a harness through Sleev with a reserved /sleev/ prefix in its base URL instead of routing headers. # URL-based routing Some tools can only be pointed at a base URL and cannot send custom request headers. Instead of adding sleev-harness and sleev-provider (or sleev-base-url) headers, you can put all the routing information in the base URL itself. Sleev recognizes a reserved /sleev// path prefix on the gateway and reads the harness and provider from it: ~/.hermes/.env ~~~ ANTHROPIC_BASE_URL=http://127.0.0.1:17321/sleev/hermes/anthropic ~~~ This is equivalent to sending sleev-harness: hermes and sleev-provider: anthropic on every request, but the harness never has to know about Sleev headers — it just uses the URL as its base. For a custom upstream not built into Sleev, put the complete upstream API URL after the harness name: ~/.hermes/.env ~~~ NOVITA_BASE_URL=http://127.0.0.1:17321/sleev/hermes/https://api.novita.ai/openai/v1 ~~~ The endpoint path the harness appends after the reserved prefix is forwarded upstream, and the prefix is removed before anything leaves the gateway. Provider credentials still travel in the request's normal Authorization or API key header. Harness names accept the same values and aliases as the sleev-harness header. --- ## Gateway bind address - Configuration | Sleev Docs URL: https://sleev.ai/docs/config/gateway-bind Description: Make the local Sleev gateway available on a trusted network. # Gateway bind address By default, the Sleev gateway listens on 127.0.0.1:17321, so only harnesses on the same machine can reach it. When a harness on another trusted machine needs to connect over your local network, run: ~~~ sleev gateway bind --host 0.0.0.0 --restart ~~~ Use --port to choose a different gateway port: ~~~ sleev gateway bind --host 0.0.0.0 --port 17324 --restart ~~~ Return to the default local-only address with: ~~~ sleev gateway bind reset --restart ~~~ Use the gateway bind address as the remote harness base URL, such as http://:17321. Only bind to 0.0.0.0 on networks you trust. --- ## Limited internet - Configuration | Sleev Docs URL: https://sleev.ai/docs/config/limited-internet Description: Configure Sleev in limited-internet environments. # Limited internet If the target machine cannot use normal browser sign-in, or Sleev should avoid sharing tracking data and usage patterns, accounts approved for offline licensing can activate the local gateway with an [offline license](/docs/offline-licensing). Note: offline licensing is currently only supported for enterprise customers. To request access, contact support at support@sleev.ai. Once licensed, harness configuration is unchanged: point it at the local gateway and send sleev-harness plus either sleev-provider or sleev-base-url. --- ## Deconfigure - Configuration | Sleev Docs URL: https://sleev.ai/docs/config/deconfigure Description: Stop routing harness traffic through Sleev. # Deconfigure To remove Sleev from this machine, sign out: ~~~ sleev auth logout ~~~ Signing out removes local auth state and gateway config, deconfigures supported harnesses, and stops the local gateway service. You can also sign out from the sleev TUI. To stop routing only one harness, disable its provider in the Config Wizard. For manual setup, remove the Sleev-specific settings shown in [Harness Setup](/docs/harness-setup). --- ## Harness Setup | Sleev Docs URL: https://sleev.ai/docs/harness-setup Description: Manual configuration examples for supported Sleev harnesses. # Harness Setup The Config Wizard can configure supported harnesses for you. Choose a harness in the navigation to edit its settings by hand or verify what sleev writes. Before you start, check that the local gateway is running: ~~~ sleev gateway status ~~~ --- ## Claude Code CLI Setup & Token Cost Optimization | Sleev Docs URL: https://sleev.ai/docs/harness-setup/claude-code Description: Configure Claude Code to use the local Sleev gateway. # Claude Code CLI Run sleev harness claude-code enable , or enable Claude Code in the sleev TUI. Both update your Claude Code settings. Restart Claude Code if it is running. To remove Sleev, run sleev harness claude-code disable . ## Manual setup Merge the env block for your OS into ~/.claude/settings.json. Create the file if it doesn't exist. ### macOS and Linux ~/.claude/settings.json ~~~ { "env": { "HTTPS_PROXY": "http://claude-code:x@127.0.0.1:17321", "NODE_EXTRA_CA_CERTS": "/.local/share/sleev/https-proxy/ca.pem", "CLAUDE_CODE_SHELL_PREFIX": "/.local/share/sleev/https-proxy/claude-code-shell-prefix.sh", "CLAUDE_CODE_DISABLE_NONSTREAMING_FALLBACK": "1", "CLAUDE_CODE_TETHER_LIVE": "false" } } ~~~ Replace with the full path of your home folder, for example /Users/you or /home/you. ~ doesn't work here. If you set XDG_DATA_HOME, the files are in $XDG_DATA_HOME/sleev/https-proxy. The gateway creates or updates both files when it starts. ### Windows %USERPROFILE%\.claude\settings.json ~~~ { "env": { "ANTHROPIC_BASE_URL": "http://127.0.0.1:17321/sleev/claude-code/anthropic", "CLAUDE_CODE_DISABLE_NONSTREAMING_FALLBACK": "1", "ENABLE_TOOL_SEARCH": "true" } } ~~~ ## Setup modes Sleev connects Claude Code in one of two modes: HTTPS proxy mode (the default on macOS and Linux) or Base URL mode (the default on Windows). The automated setup command (sleev harness claude-code enable) configures the recommended mode for your operating system automatically. If you set up a mode by hand that isn't the default for your OS, disable Claude Code in sleev and then enable it again to change it back to the default. ### HTTPS proxy mode In HTTPS proxy mode, Claude Code routes its HTTPS traffic through the Sleev gateway proxy without needing to set ANTHROPIC_BASE_URL. This keeps features like Remote Control and mobile push notifications enabled. Sleev decrypts traffic to api.anthropic.com, optimizes model requests, and passes all other traffic through unchanged. Key details: - Certificate trust: The gateway decrypts HTTPS traffic to api.anthropic.com, so NODE_EXTRA_CA_CERTS tells Claude Code to trust Sleev's local root CA (ca.pem). - Existing proxies: Sleev sets HTTPS_PROXY only for Claude Code, not system-wide. If you already use an HTTPS proxy with Claude Code, Sleev replaces it, and the gateway doesn't forward traffic to your proxy. - Process isolation: CLAUDE_CODE_SHELL_PREFIX unsets Sleev proxy variables before running shell tool commands, status lines, and stdio MCP servers on macOS and Linux. This prevents most child commands from inheriting Sleev's proxy and certificate settings, though commands executed without a shell or as exec-form hooks still retain them. Claude Code runs PowerShell commands without the shell prefix, and stdio MCP servers on Windows cannot execute the shell script outside Git Bash. This is why Base URL mode is the default on Windows. ### Base URL mode In Base URL mode, Claude Code routes Anthropic model requests directly to the Sleev gateway via ANTHROPIC_BASE_URL. All other connections bypass the gateway and connect directly. Key details: - No certificate required: Claude Code sends model requests to the gateway over plain HTTP on 127.0.0.1, so you don't need to trust a local root certificate. - Child processes: Subprocesses started by Claude Code, including shell commands, hooks, and stdio MCP servers, inherit ANTHROPIC_BASE_URL. Any child scripts or tools that use the Anthropic SDK will also route their API requests through the Sleev gateway. - Disabled features: Setting ANTHROPIC_BASE_URL causes Claude Code to turn off features such as Remote Control (including mobile push notifications) and server-managed settings. To use Base URL mode on macOS or Linux instead of the proxy, first run sleev harness claude-code disable to remove the HTTPS proxy settings. Then add the [Windows settings block](#windows) above to ~/.claude/settings.json. --- ## Claude Desktop Setup & Token Cost Optimization | Sleev Docs URL: https://sleev.ai/docs/harness-setup/claude-desktop Description: Configure Claude Desktop with Sleev. # Claude Desktop Claude Desktop can route through Sleev using its third-party inference mode and the claude-desktop-cred helper. For full details, see the official [Claude Desktop Gateway guide](https://code.claude.com/docs/en/llm-gateway-connect#desktop-app) and [Configuration Reference](https://claude.com/docs/third-party/claude-desktop/configuration) . ## Requirements In order to use Sleev with Claude Desktop, you must: - Be logged in with the Claude Code CLI using your Claude subscription: claude login - Retrieve the credential helper path: sleev --credential-helper - Set the Claude Code gateway base URL: http://127.0.0.1:17321/sleev/claude-code/anthropic - Open Help → Troubleshooting → Enable Developer Mode… - Open Developer → Configure Third-Party Inference… - Choose Gateway and Helper script, then paste Sleev’s base URL. - Paste the helper script path, set its TTL, and test it. - Apply, Claude Desktop now uses Sleev. --- ## Codex / ChatGPT Setup & Token Cost Optimization | Sleev Docs URL: https://sleev.ai/docs/harness-setup/codex Description: Configure the ChatGPT app or Codex CLI to use the local Sleev gateway. # Codex / ChatGPT The ChatGPT app and the Codex CLI use the same setup. They pick their upstream by named provider in ~/.codex/config.toml, which takes two edits. Put the first part at the top of the file: ~/.codex/config.toml ~~~ model_provider = "sleev" ~~~ Then put the second part at the bottom: ~/.codex/config.toml ~~~ [model_providers.sleev] name = "Sleev" base_url = "http://127.0.0.1:17321/v1" wire_api = "responses" supports_websockets = true requires_openai_auth = true [model_providers.sleev.http_headers] "sleev-provider" = "openai" "sleev-harness" = "chatgpt" ~~~ Set sleev-provider to openai when you use an OpenAI API key, or codex when you sign in with your ChatGPT account. --- ## OpenCode Setup & Token Cost Optimization | Sleev Docs URL: https://sleev.ai/docs/harness-setup/opencode Description: Configure OpenCode providers to use the local Sleev gateway. # OpenCode OpenCode manages each provider directly, so override the connection per provider in ~/.config/opencode/opencode.json. Deep-merge the block below under provider and compaction. This example configures Anthropic. Repeat the provider block for each provider you use. ~/.config/opencode/opencode.json ~~~ { "compaction": { "prune": false }, "provider": { "anthropic": { "options": { "baseURL": "http://127.0.0.1:17321", "headers": { "sleev-provider": "anthropic", "sleev-harness": "opencode" } } } } } ~~~ This sends the Anthropic provider through the local Sleev gateway. Setting compaction.prune to false leaves context management to Sleev. OpenCode 2 (currently in beta) uses a new provider config shape. Use providers instead of provider, put baseURL under settings, and put headers directly on the provider. OpenCode 2 does not use compaction.prune. ~/.config/opencode/opencode.json ~~~ { "providers": { "anthropic": { "settings": { "baseURL": "http://127.0.0.1:17321" }, "headers": { "sleev-provider": "anthropic", "sleev-harness": "opencode" } } } } ~~~ --- ## Pi Setup & Token Cost Optimization | Sleev Docs URL: https://sleev.ai/docs/harness-setup/pi Description: Configure Pi providers to use the local Sleev gateway. # Pi Pi reads provider overrides from ~/.pi/agent/models.json. If PI_CODING_AGENT_DIR is set, use models.json in that directory instead. Merge the block below into your config, preserving the provider's API key and other headers. Repeat it for each provider you use. ~/.pi/agent/models.json ~~~ { "providers": { "anthropic": { "baseUrl": "http://127.0.0.1:17321", "headers": { "sleev-provider": "anthropic", "sleev-harness": "pi" } } } } ~~~ --- ## Grok Build Setup & Token Cost Optimization | Sleev Docs URL: https://sleev.ai/docs/harness-setup/grok-build Description: Configure Grok Build to use the local Sleev gateway. # Grok Build Grok Build reads ~/.grok/config.toml. Merge the blocks below into that file. The endpoint routes built-in Grok models through Sleev, while the extra headers provide Sleev routing and the auth values Grok's servers expect. ~/.grok/config.toml ~~~ [endpoints] cli_chat_proxy_base_url = "http://127.0.0.1:17321" [models.extra_headers] sleev-harness = "grok-build" sleev-provider = "grok" X-XAI-Token-Auth = "xai-grok-cli" x-authenticateresponse = "authenticate-response" ~~~ With this in place, requests bill to your Grok subscription sign-in. ## xAI API billing To use a pay-as-you-go xAI API key instead, remove the subscription blocks and add this custom provider. Paste your key from console.x.ai into api_key, or use env_key = "XAI_API_KEY". Sleev sends the key per request and never stores it. ~/.grok/config.toml ~~~ [model_providers.sleev-xai] base_url = "http://127.0.0.1:17321/v1" api_backend = "responses" api_key = "" [model_providers.sleev-xai.extra_headers] sleev-harness = "grok-build" sleev-provider = "xai" ~~~ Reference the provider from each model you want routed through Sleev: ~/.grok/config.toml ~~~ [model."grok-4.6"] model_provider = "sleev-xai" ~~~ Note: Sleev serves Grok Build on the Responses API only. Do not set api_backend to "messages" or "chat_completions". --- ## Hermes Agent Setup & Token Cost Optimization | Sleev Docs URL: https://sleev.ai/docs/harness-setup/hermes-agent Description: Configure Hermes Agent providers to use the local Sleev gateway. # Hermes Agent Route a Hermes provider through Sleev by setting its base URL variable in $HERMES_HOME/.env. ## Providers built into Sleev These providers are built into Sleev and can use Hermes base URL variables directly. See all [providers supported by Sleev](/docs/config/providers). Set the matching Hermes base URL variable to its Sleev route, and add only the providers you use to $HERMES_HOME/.env: ~/.hermes/.env ~~~ # Add only the providers you use. ANTHROPIC_BASE_URL=http://127.0.0.1:17321/sleev/hermes/anthropic OPENAI_BASE_URL=http://127.0.0.1:17321/sleev/hermes/openai OPENROUTER_BASE_URL=http://127.0.0.1:17321/sleev/hermes/openrouter GEMINI_BASE_URL=http://127.0.0.1:17321/sleev/hermes/google KIMI_BASE_URL=http://127.0.0.1:17321/sleev/hermes/kimi-coding XIAOMI_BASE_URL=http://127.0.0.1:17321/sleev/hermes/mimo ALIBABA_CODING_PLAN_BASE_URL=http://127.0.0.1:17321/sleev/hermes/bailian-token-plan GLM_BASE_URL=http://127.0.0.1:17321/sleev/hermes/zai-coding-plan OPENCODE_ZEN_BASE_URL=http://127.0.0.1:17321/sleev/hermes/opencode OPENCODE_GO_BASE_URL=http://127.0.0.1:17321/sleev/hermes/opencode-go ~~~ ## Providers not built into Sleev For a Hermes provider not listed above, route its complete upstream URL through Sleev. If Hermes exposes a base URL variable for it, use the [Hermes provider reference](https://hermes-agent.nousresearch.com/docs/reference/environment-variables#llm-providers) to find the variable and default URL, then put that URL after /sleev/hermes/. For example: ~/.hermes/.env ~~~ # Put the provider's full upstream base URL after /sleev/hermes/. NOVITA_BASE_URL=http://127.0.0.1:17321/sleev/hermes/https://api.novita.ai/openai/v1 MINIMAX_BASE_URL=http://127.0.0.1:17321/sleev/hermes/https://api.minimax.io/anthropic ~~~ ## Custom upstreams If Hermes does not expose a base URL variable for an endpoint, define a named custom provider. Put the complete upstream API base after /sleev/hermes/: ~/.hermes/config.yaml ~~~ providers: sleev-custom: api: "http://127.0.0.1:17321/sleev/hermes/https://provider.example.test/v1" key_env: CUSTOM_PROVIDER_API_KEY transport: chat_completions model: provider: custom:sleev-custom default: your-model-id ~~~ Replace the model and URL placeholders, and store CUSTOM_PROVIDER_API_KEY in $HERMES_HOME/.env. Use chat_completions, codex_responses, or anthropic_messages for the matching endpoint format. For Anthropic, omit a terminal /v1. ## Codex OAuth Codex OAuth requires a manual edit because Hermes currently ignores HERMES_CODEX_BASE_URL in credential-pool paths. See [Hermes issue #40913](https://github.com/NousResearch/hermes-agent/issues/40913) . Sleev does not modify auth.json because Hermes owns the rotating OAuth tokens in that file. Warning: This is a workaround, not a stable configuration interface. Hermes manages and may regenerate auth.json; signing in again or upgrading Hermes can remove the Sleev URL. Check the pool URL again after either. Stop Hermes, then open $HERMES_HOME/auth.json. Keep every existing field in each credential_pool.openai-codex entry and change only its base_url. This fragment shows the location: ~/.hermes/auth.json (relevant fields) ~~~ { "credential_pool": { "openai-codex": [ { "base_url": "http://127.0.0.1:17321/sleev/hermes/codex" } ] } } ~~~ Do not replace the entry or change its access or refresh tokens. If the file also contains a legacy providers.openai-codex entry, remove it; Hermes uses it to restore the direct OpenAI URL. --- ## Premium Requests | Sleev Docs URL: https://sleev.ai/docs/premium-requests Description: What counts as a Premium request, which models are currently premium, monthly quotas, and what happens when a quota runs out. # Premium requests Sleev meters usage based on Premium requests. Most traffic through your gateway is completely free: a request only consumes quota when it targets a model on Sleev's Premium list and Sleev saves tokens on the request. ## What counts as a Premium request A request passing through the Sleev gateway counts as a Premium request only when both conditions are met: - Premium model: It targets a model included in Sleev's active [Premium model list](/docs/premium-models). - Token savings achieved: Sleev saves more than 0 tokens on the request. ## Premium models Sleev maintains a curated list of frontier models designated as Premium. See the full list on the [Premium models](/docs/premium-models) page. All other models, including other provider models, open-weight models, and custom self-hosted endpoints, are completely free to optimize with Sleev. ## Monthly quotas & tracking Every Sleev installation includes a monthly allowance of Premium requests. In account-free mode (using Sleev without an account), you get 500 free Premium requests per month. Creating a free account and signing in increases your monthly allowance to 2,000 free Premium requests per month. [Paid plans](/pricing) offer many more Premium requests per month. You can monitor your current usage, remaining quota, and reset dates in real time on the [dashboard](/dashboard) or directly inside the sleev terminal app. Quotas reset automatically at the start of each calendar month at 00:00 UTC, regardless of your billing date. Unused request allowances do not roll over between cycles. ## What happens when your quota is reached When your quota is reached, Premium requests continue to your provider without further optimization. Existing session optimizations remain in place. Models outside the Premium list keep full optimization regardless of quota status. Optimization for Premium models resumes automatically when your quota resets at the start of the next calendar month, or as soon as you upgrade your plan. --- ## Premium Models | Sleev Docs URL: https://sleev.ai/docs/premium-models Description: The model families that count as Premium with Sleev, grouped by provider. # Premium models Sleev maintains a curated set of frontier model families designated as Premium. Requests to these models count against your monthly [Premium request](/docs/premium-requests) quota when Sleev saves tokens on them. Models outside these Premium families are completely free to optimize with Sleev. Provider Model Family Loading models… Enable JavaScript to load the current model catalog. Try again Every model in a listed family counts as Premium, including older ones. For example, Claude Opus 4.8, GPT-5.4 Pro, and Gemini 2.5 Pro are Premium models. --- ## Offline Licensing | Sleev Docs URL: https://sleev.ai/docs/offline-licensing Description: Use offline licenses to run Sleev on licensed machines with limited internet access. # Offline licensing Sleev accounts approved for offline licensing can run the local gateway almost entirely offline. When using an offline license, your local gateway does not report any telemetry to our servers. The only internet connectivity required is downloading or updating Sleev, and generating licenses in your account dashboard to set up local installations. Note: offline licensing is currently only supported for enterprise customers. To request access, contact support at support@sleev.ai. --- ## When to use it - Offline Licensing | Sleev Docs URL: https://sleev.ai/docs/offline-licensing/when-to-use Description: When to use Sleev offline licensing in limited-internet environments. # When to use it Use offline licensing for licensed machines that cannot rely on browser sign-in or regular online account checks. It is useful for restricted networks, lab machines, jump boxes, and other limited-internet environments where Sleev should not share tracking data or usage patterns. Offline licensing does not make your model provider offline. Your harness still needs whatever access and credentials its upstream provider requires. Installing or updating Sleev, creating a license, and renewing a license also require online access from whichever machine or browser performs those steps. --- ## Create a license - Offline Licensing | Sleev Docs URL: https://sleev.ai/docs/offline-licensing/create Description: Create and apply a Sleev offline license. # Create a license - On the machine that will run Sleev, install the sleev CLI and run: ~~~ sleev auth license request ~~~ - Sign in to the website with the licensed account and open [Dashboard, then Licenses](/dashboard/licenses). - Paste the slv_act_... request, add an optional label, and create the offline license. - Copy the returned slv_lic_... license into the waiting CLI prompt. If you closed it, apply the license on the same machine with: ~~~ sleev auth license apply ~~~ --- ## Renew - Offline Licensing | Sleev Docs URL: https://sleev.ai/docs/offline-licensing/renew Description: Renew a Sleev offline license before it expires. # Renew Offline licenses expire. Renew before the expiration shown by the CLI or dashboard. ~~~ sleev auth license status ~~~ When the license is inside its renewal window, generate a renewal request on the same licensed machine: ~~~ sleev auth license renew-request ~~~ Paste the generated slv_renew_... request into the matching license in the dashboard, then apply the returned slv_lic_... replacement on the same installation. ~~~ sleev auth license apply ~~~ --- ## Limits and recovery - Offline Licensing | Sleev Docs URL: https://sleev.ai/docs/offline-licensing/limits Description: Offline license limits, recovery, updates, and removal. # Limits and recovery - One offline license is bound to one local Sleev installation. - For multiple machines, generate one activation request per machine. - Copying only the slv_lic_... string to another machine will not activate it there. - If a machine is lost or needs to be rehosted, contact support. - Keep Sleev updated so it includes the current offline-license public keys. ~~~ sleev upgrade ~~~ To remove the active license while preserving the local offline identity for future requests, run: ~~~ sleev auth license clear ~~~ If applying a license succeeds but the gateway does not restart cleanly, restart it manually: ~~~ sleev gateway restart ~~~ --- ## Privacy - Offline Licensing | Sleev Docs URL: https://sleev.ai/docs/offline-licensing/privacy Description: How Sleev handles data during licensed offline use. # Privacy During licensed offline use, the gateway verifies the signed license locally. Request metrics used by the local UI are recorded locally on the machine. No gateway telemetry is sent to Sleev's control-plane services during licensed day-to-day gateway use. Configured-provider requests and supported token-counting calls still apply. See the [offline licensing privacy policy](/privacy#offline-licensing) for details. --- ## CLI Reference | Sleev Docs URL: https://sleev.ai/docs/cli-reference Description: A compact reference for sleev CLI commands. # CLI Reference These are the commands exposed by sleev. Run sleev --help or the command's -h flag when available for exact usage. ## Setup and status Guide: [Quickstart setup](/docs/quickstart/setup) - sleev [--version] — Open the TUI, or print the CLI version with --version. - sleev setup [--yes] — Register the machine, install the gateway if missing, and start Sleev. Setup does not open a browser. --yes accepts the guest terms without prompting. - sleev --credential-helper — Install only the latest Claude Desktop credential helper and print its path, without setting up the gateway. - sleev status [--config] — Print gateway status as JSON. --config includes the config. - sleev upgrade [--yes] [version] — Update the sleev CLI and gateway together. --yes skips the confirmation for updates that break running sessions. ## Harnesses Guide: [Harness Setup](/docs/harness-setup) - sleev harness — Configure harness/provider connections. - sleev harness list — List supported harnesses, detection status, and enabled providers. - sleev harness list — List provider IDs and state for one harness. - sleev harness enable [provider] — Route a provider through Sleev. Claude Code and Codex select their provider automatically. - sleev harness disable [provider|--all] — Remove one or all Sleev-owned routes for a harness. ## Account - sleev auth — Open the account menu, including browser, device-code, and browser-link sign-in options. - sleev auth login [--no-open|--headless|--device] — Sign in through the browser by default. --headless and --device select device-code login for machines without a local browser. --no-open prints the browser-login link instead of opening it; this flow still uses a local callback. Choose at most one flag. - sleev auth status — Show the current auth state. - sleev auth logout [--yes] — Sign out and continue in Guest mode. --yes accepts the required guest terms when leaving an offline license. - sleev auth verify — Check the current account token. - sleev usage [--json] — Show Premium request usage, attribution, and reset date for the active account or organization. --json prints it as JSON. ## Organizations - sleev org — Open the attribution menu. Choose attribution offers your personal account and available organizations. Organization usage still appears on your personal dashboard. - sleev org status — Show the current attribution. - sleev org list — Show the organizations you can attribute usage to. - sleev org set — Attribute usage to an organization you belong to. - sleev org clear — Attribute usage to your personal account. ## Offline licenses Guide: [Offline Licensing](/docs/offline-licensing) - sleev auth license — Open the offline license menu. - sleev auth license request — Create an offline activation request. - sleev auth license apply — Activate an offline license. - sleev auth license renew-request — Create an offline license renewal request. - sleev auth license status — Show offline license status. - sleev auth license clear — Remove the active offline license. ## Gateway Guide: [Gateway bind address](/docs/config/gateway-bind) - sleev gateway — Open the gateway menu. - sleev gateway status — Show gateway status. - sleev gateway start|restart|stop — Manage the local gateway. - sleev gateway bind --host [--port ] [--restart] — Change the gateway listen address. - sleev gateway bind reset [--restart] — Reset the listen address to local defaults. ## Logs and feedback - sleev feedback [--logs [--yes]] — Send feedback to Sleev from an interactive terminal. --logs uploads diagnostic logs for support instead; conversation data is redacted before logs are saved. --yes skips the confirmation. ## Experimental - sleev experimental — Open the experimental features menu. - sleev experimental cache-breakpoints [status|enable|disable] — Smarter cache breakpoints for Claude models on non-Anthropic upstreams. - sleev experimental cache-warming [status|enable|disable] — Keep idle Claude prompt caches warm (default in Claude Code). - sleev experimental opencode-claude-auth [status|enable|disable] — OpenCode Claude Auth compatibility, for use with the opencode-claude-auth plugin. - sleev experimental request-logs [status|enable|disable] — Show, enable, or disable local outgoing request logging. Run without a subcommand to toggle it, or use the experimental checkbox menu. Disabled by default. --- ## FAQ | Sleev Docs URL: https://sleev.ai/docs/faq Description: Answers to common questions about Sleev savings, privacy, cancellation, supported harnesses, provider credentials, licensing, and feedback. # Frequently asked questions Common questions about Sleev. ## Can I use Sleev at work? Individual solo commercial, freelance, contractor, and client work is allowed when one person uses Sleev as their own tool. Organization evaluation is limited to 30 consecutive days, one evaluator, one local machine, and either one account or one unregistered installation. The period applies to the organization as a whole and does not restart when people, accounts, devices, or installations change. Regular employer/team/company use after that requires a pilot, enterprise subscription, or written agreement. For company evaluations, contact contact@sleev.ai. ## How does Sleev work? Sleev runs as a local gateway between your coding tool and your model provider. It optimizes session context to reduce token usage, weighing the expected savings against provider-cache costs when deciding when to optimize. ## Can Sleev run in limited-internet environments? Yes, for accounts approved for offline licensing. Offline licensing lets the local gateway verify a signed license on the machine instead of relying on normal account sign-in, regular online account checks, or usage-pattern uploads to Sleev. The model provider itself may still require network access from that environment. See the [offline licensing docs](/docs/offline-licensing) for the activation and renewal flow. ## Do you store my conversations? No. Sleev does not send your conversation history, prompts, code, model responses, tool content, or summaries to our servers, including in optional diagnostic uploads. Sleev receives account and operational information to manage access, billing, usage, and product reliability. ## Do you collect provider credentials? No. Your provider credential remains with your harness or provider configuration. The harness sends it to Sleev with each proxied request, and Sleev forwards it upstream unchanged without storing it. ## How do I stop using Sleev? Remove the Sleev-specific entries from your harness's config file and it will go back to direct provider calls. The [Deconfigure](/docs/config/deconfigure) section shows what to remove. This stops the harness from using Sleev, but a local gateway process may still be running. To stop and remove the gateway too, open the sleev TUI and sign out. This will also deconfigure any detected harness config file. ## Which providers and harnesses are supported? Sleev can route requests to supported providers depending on the harness you configure. Claude Code, OpenCode, Codex, Pi, Grok Build, and Hermes Agent are documented in the [configuration docs](/docs/config). ## How can I get in contact or give feedback? Email support@sleev.ai or join the [Sleev Discord](https://discord.gg/W36rxJswbG) to ask questions, report issues, or share feedback with the team. Every way to reach us is listed on the [contact page](/contact). ## Is there a service level agreement (SLA)? We do not offer a formal SLA unless a written enterprise agreement says otherwise. --- ## Privacy Policy | Sleev URL: https://sleev.ai/privacy Description: What Sleev collects, what stays local, and how we use it. # Privacy Policy Last updated: September 9, 2026 Effective date: 2026-09-09 This Privacy Policy describes how Sleev Labs Inc. (“Sleev,” “we,” or “us”) collects, uses, and discloses personal data, and the rights and choices available to you. It applies to our website, software, hosted services, and support. Sleev’s gateway processes coding sessions locally on your device. We do not collect your prompts, code, conversation history, model responses, tool content, or provider API keys. We collect and use account and operational information to provide and administer our services, process billing, understand usage, evaluate and improve Sleev’s features and performance, provide support, and maintain security. Sleev Labs Inc. is responsible for the personal data described in this policy. We do not sell personal data or use customer data to train AI models. Where we process personal data on an organization’s behalf under a [Data Processing Addendum](/dpa), that agreement governs the processing. ## Local processing The Sleev gateway processes coding sessions and stores its database and operational files on your device. Prompts, code, file contents, conversation history, model responses, tool inputs and outputs, and content generated during context optimization are not sent to our servers. To carry out your requests, the local gateway connects to the AI provider or endpoint you configure. Your provider credentials are used locally to authenticate those requests; Sleev’s servers do not receive them. Your chosen provider handles the requests it receives under its own terms and privacy policy. ## Information we collect and use - Account and billing information, such as your email, account and team details, subscription, and payment status, to manage access and provide the service. These details come from you, your sign-in provider, or your organization administrator. Stripe processes payments; Sleev receives the billing and subscription records needed to manage your plan. - Usage and technical information, such as usage totals, performance metrics, software versions, and installation and session identifiers, to operate and improve Sleev, manage licenses, troubleshoot problems, and prevent abuse. This includes metadata about individual requests and sessions, not prompt content, which may be associated with your account or installation. - Information you send us, such as support messages and optional diagnostic logs, to answer questions and troubleshoot problems. Guest installations use a locally generated, one-way hash to recognize the installation, including after a reinstall. We do not receive the underlying machine identifier. ## Optional diagnostics You can choose to upload diagnostic logs through the Sleev CLI or TUI for support, debugging, security, and reliability. These uploads contain operational information, such as device names, file paths, software and system details, and diagnostic events. Diagnostic uploads do not include your prompts, code, conversation history, model responses, tool content, summaries, or provider API keys. ## Offline licensing In offline-license mode, the gateway validates its license locally and does not send gateway usage telemetry to Sleev. Metrics shown in the local UI stay on your machine. Optional diagnostic uploads remain available. Creating, renewing, or supporting an offline license may require account, billing, support, and license-administration information. See [Offline Licensing](/docs/offline-licensing/privacy) for details. ## Disclosure of personal data We use service providers for authentication, hosting, databases, storage, payments, email, support, and security. They process information needed to provide those services. Our [Subprocessors Directory](/subprocessors) lists these providers. We may also disclose information to meet legal obligations, protect the rights and safety of Sleev and its users or others, or carry out a corporate transaction such as a merger, financing, acquisition, reorganization, or sale of assets. ## Retention and deletion - Diagnostic uploads are retained for no longer than seven days and are removed when you delete your account. - Account profile and sign-in information is removed when account deletion completes. Account deletion also revokes Sleev credentials and ends access to the account. - Usage, installation, and operational records may be retained after account deletion as reasonably necessary to operate and improve Sleev, measure usage and performance, maintain security, and investigate problems. Retained records may remain associated with internal account or installation identifiers. Uninstalling the software does not delete hosted records. - Billing, security, backup, and legal records, including records of your acceptance of our terms and related notices, may remain after account deletion where needed for legal obligations, dispute resolution, security, or enforcement of our agreements. We retain personal data only for as long as necessary for the purposes described in this policy. Retention periods depend on the type of information, how it is used, and applicable legal requirements, subject to your privacy rights. Where an applicable DPA governs the processing, its retention and deletion requirements apply. Deleting your Sleev account does not remove files on your machine or information held by your AI provider. You control local files; your provider’s policy governs its records. For questions about retained records or to request deletion, including for a guest installation, contact support@sleev.ai. ## Your privacy rights Depending on where you live, you may have the right to access, correct, delete, or export personal data we hold about you, and to restrict or object to certain uses. Email support@sleev.ai to make a request. We may need to verify your identity before completing it. Where we rely on legitimate interests, you may object on grounds relating to your circumstances, subject to applicable law. Where we rely on consent, you can withdraw it at any time; withdrawal does not affect processing already carried out lawfully. You may also raise a complaint with your local data protection authority. ## Legal bases for processing Where data protection law requires a legal basis, we rely on: - Our contract with you to provide Sleev, authenticate access, administer accounts and licenses, and handle billing. - Legitimate interests to understand and improve product performance, provide support, maintain security, prevent abuse, and keep records of our agreements. - Legal obligations to meet applicable legal requirements. - Consent where we ask for your permission and rely on it for a particular use. ## Security and international processing We use administrative, technical, and organizational safeguards to protect personal data. Sleev Labs Inc. is based in the United States. We and our service providers may process personal data in the United States and other countries, whose privacy laws may differ from those where you live. Where required by applicable law, we rely on adequacy decisions or appropriate safeguards, including Standard Contractual Clauses, for international transfers of personal data. Contact support@sleev.ai for information about the safeguards applicable to your data and how to obtain a copy. ## Children Sleev is not directed to children under 13, and we do not knowingly collect personal data from them. If you believe a child has provided personal data to us, please contact us so we can address it. ## Changes and contact We may update this policy as Sleev evolves. We will publish revisions here and update the date at the top of the page. For privacy questions or requests, contact Sleev Labs Inc. at support@sleev.ai. --- ## Terms of Service | Sleev URL: https://sleev.ai/terms Description: Terms for using Sleev. # Terms of Service Version: 2026-09-01 Effective date: 2026-09-01 These Terms are an agreement between you and Sleev Labs Inc. (“Sleev Labs,” “we,” “us,” or “our”) and govern your use of Sleev’s website, accounts, hosted services, local software, and related services. By using Sleev, you agree to these Terms. If you do not agree, do not use Sleev. The [EULA](/eula) governs your license to install and use the local Sleev software. The EULA controls local-software license grants, license categories, and installed-software restrictions. These Terms control accounts, hosted services, and the overall relationship between you and Sleev Labs. If you have a pilot agreement, enterprise agreement, order form, or other agreement expressly entered into with Sleev Labs, that agreement controls to the extent it conflicts with these Terms or the EULA. ## 1. Eligibility and accounts Sleev is not directed to children. Users under 13 may not use Sleev. Users under the age of majority where they live may use Sleev only with permission from a parent or legal guardian. If you use Sleev on behalf of an organization, you represent that you have authority to bind that organization. If you do not have that authority, you may use Sleev only for your own permitted individual use or limited evaluation use. If you create an account, you must provide accurate information, keep it current, protect your credentials, and notify us at support@sleev.ai if you believe your account or credentials have been compromised. You are responsible for activity through your account, local installation, provider credentials, upstream provider accounts, and systems you connect to Sleev. ## 2. Services and local software Sleev provides software for running a local LLM gateway on your machine. The gateway processes requests between supported harnesses and the LLM providers or endpoints you configure. Sleev’s hosted services support accounts, gateway connections, usage reporting, diagnostics, licensing, updates, and related features. Subject to these Terms and any plan or agreement that applies to you, Sleev Labs grants you a limited, non-exclusive, non-transferable right to access and use the hosted services made available to you. The EULA defines individual use, organization evaluation, and organization use of the local software. Any organization use beyond the EULA’s self-serve evaluation limits requires a paid pilot, enterprise subscription, or written agreement with Sleev Labs. ## 3. Acceptable use You may use Sleev only in a lawful manner, only where you have the right to use it, and only within the scope allowed by these Terms, the EULA, and any other agreement that applies to you. You must not use Sleev to violate the rights of others, access or attack systems without authorization, bypass provider restrictions, interfere with or disrupt Sleev, evade authentication or usage limits, abuse upstream providers, distribute malicious code, or violate export-control or sanctions laws. You may not sell, sublicense, or make Sleev’s hosted services available to others except as expressly allowed by your plan or written agreement. You are responsible for securely configuring the local gateway. Do not expose it to untrusted networks unless you understand and accept the risk. Local-software installation rights and restrictions, including rules for redistribution, reverse engineering, shared use, and benchmark publication, are in the EULA. ## 4. Providers and data Your configured LLM provider or endpoint receives the model requests sent from the local gateway. Your agreement with that provider governs its services and charges. Sleev Labs is not your model provider and does not control provider output, availability, pricing, data handling, policy enforcement, or API changes. Sleev’s optimization pipeline may make additional provider requests. The [Privacy Policy](/privacy) explains what remains on your machine, what Sleev Labs receives, when content is sent to providers, and how we handle personal data. ## 5. Plans, subscriptions, fees, and usage limits Paid self-serve subscriptions require a Sleev account. Enterprise licenses and pilots are governed by the applicable written agreement or order form. Purchasing a paid individual plan does not expand the organization-use rights granted by the EULA. Before you subscribe, we will show the price, currency, billing interval, applicable taxes, recurring nature of the subscription, plan limits, and any additional transaction terms. By purchasing a recurring subscription, you authorize Sleev Labs and our payment processor to charge the disclosed amounts in advance for each billing period. Your subscription automatically renews for the same interval until you cancel it. You may cancel at any time through the available account billing settings or payment portal. If you cancel before your next renewal, you will not be charged for another billing period. Unless we state otherwise when you subscribe, your paid benefits continue through the end of the current paid period, after which your account returns to any free access for which it is eligible. Refunds or credits may be available based on the circumstances, including billing errors or service failures, and where required by law. You may request one by contacting support@sleev.ai. Any additional refund rights shown when you subscribe also apply. If a payment fails or remains overdue, we or our payment processor may retry the charge, and we may suspend paid features or return your account to an eligible free plan. We will give at least 30 days’ advance notice of an increase to your recurring subscription price. An increase takes effect no earlier than your next renewal after that notice, and you may cancel before it takes effect. Plan changes, trials, and promotions may have additional terms presented when you select them. Using Sleev may create usage on your configured provider, including requests made as part of Sleev’s optimization pipeline. Provider charges are separate from Sleev fees, are billed by your provider, and are your responsibility. Features that do not require an account are subject to the usage limits displayed by Sleev, and we may change those limits. When the applicable limit is exhausted, Sleev’s optimization pipeline is unavailable until the limit resets, while the local gateway may continue forwarding ordinary requests to your configured provider. ## 6. Ownership and submissions Sleev Labs and its licensors own Sleev’s hosted services, local software, websites, documentation, branding, and related intellectual property. These Terms grant only the rights expressly stated here. The EULA grants the applicable rights in the local software. You retain your rights in material you voluntarily submit to Sleev Labs, such as support messages, feedback, and diagnostic uploads. You give Sleev Labs the rights needed to receive, host, reproduce, and process that material to operate, secure, support, and improve Sleev. This does not grant Sleev Labs rights in provider traffic that you do not submit to us. You may say that you use Sleev. You may not use Sleev’s name, logos, or branding in a way that suggests endorsement, partnership, resale, official status, or ownership without our written permission. ## 7. Availability, suspension, and termination Self-serve users receive no guaranteed support, uptime, service-level agreement, or response time. Enterprise and pilot customers receive only the commitments stated in their written agreement or order form. We may change, suspend, or discontinue features. We may suspend or terminate access to accounts or hosted services for abuse, illegal use, security risk, nonpayment, violation of these Terms, or use outside the allowed license scope. You may stop using Sleev at any time. Account deletion and data retention are described in the Privacy Policy. Ending hosted access does not delete files stored on your machine or, by itself, terminate rights granted under a separate offline or enterprise license. Ending a local-software license is governed by the EULA or the applicable written agreement. Sections that by their nature should continue after termination remain in effect, including ownership, payment obligations, disclaimers, liability limits, governing law, and general terms. ## 8. Disclaimers To the fullest extent permitted by law, Sleev is provided “as is” and “as available,” without warranties of any kind. Sleev Labs disclaims implied and statutory warranties, including warranties of merchantability, fitness for a particular purpose, title, non-infringement, availability, accuracy, and uninterrupted or error-free operation. We do not guarantee provider compatibility, model behavior, specific results, lower provider charges, token savings, or the accuracy or safety of provider output or agent actions. You are responsible for deciding whether Sleev and any connected provider are appropriate for your use. Enterprise and pilot customers receive only the warranties stated in their written agreement or order form. ## 9. Liability and user responsibility To the fullest extent permitted by law, Sleev Labs will not be liable for indirect, incidental, special, consequential, exemplary, or punitive damages, or for lost profits, lost revenue, lost data, business interruption, or replacement services. For free self-serve and paid individual self-serve use, Sleev Labs’ total aggregate liability arising out of or relating to Sleev, these Terms, or the EULA is limited to the greater of USD $100 or the fees you paid to Sleev Labs during the six months immediately before the event giving rise to the claim. This cap applies across all claims and legal theories and is not cumulative between these Terms and the EULA. For enterprise and pilot customers, liability limits are defined by the applicable written agreement or order form. Nothing in these Terms limits liability that cannot legally be limited. You are responsible for claims and losses caused by your unlawful use, misuse, violation of these Terms or the EULA, or infringement of another person’s rights. ## 10. Changes to Sleev and these Terms We may change Sleev or these Terms as the product evolves. For users with a Sleev account, we will give notice of material legal changes by email, website notice, or another appropriate product notice. Those users may be asked to expressly accept updated Terms or may accept them through continued use after the stated effective date. For users without a Sleev account, we provide notice by posting the revised Terms with a stated effective date. By continuing to use Sleev on or after that date, you agree to the revised Terms. If you do not agree, you must stop using and uninstall Sleev. Minor edits, formatting changes, clarifications, and other non-material changes do not require renewed acceptance. ## 11. General terms These Terms are governed by Delaware law, without regard to conflict-of-law rules. The courts located in Delaware have exclusive jurisdiction and venue for disputes relating to these Terms or Sleev, except where applicable consumer law gives you the right to bring a dispute elsewhere or an applicable written agreement specifies different terms. You may not assign these Terms without our written permission. We may assign these Terms in connection with a merger, acquisition, reorganization, or sale of assets. If any provision of these Terms is found unenforceable, the remaining provisions will remain in effect. If we do not enforce a provision, that does not waive our right to do so later. These Terms, the EULA, and any applicable plan or ordering terms are the entire agreement between you and Sleev Labs for self-serve use of Sleev, except for any other agreement expressly entered into with Sleev Labs. Headings are for convenience only. For questions or legal notices, email support@sleev.ai. --- ## EULA | Sleev URL: https://sleev.ai/eula Description: License terms for installing and running the local Sleev software. # EULA Version: 2026-09-01 Effective date: 2026-09-01 These Software License Terms (the “EULA”) are an agreement between you and Sleev Labs Inc. (“Sleev Labs,” “we,” “us,” or “our”) and govern the Sleev CLI, TUI, local gateway, updater, install scripts, packaged binaries, configuration tools, documentation distributed with them, and related local components (the “Software”). By installing or using the Software, you agree to this EULA. If you do not agree, do not install or use the Software. The [Terms of Service](/terms) govern accounts, hosted services, billing, configured providers, and the overall relationship between you and Sleev Labs. This EULA controls local-software license grants, license categories, and installed-software restrictions. The [Privacy Policy](/privacy) explains how Sleev handles personal data. If you have a pilot agreement, enterprise agreement, order form, or other agreement expressly entered into with Sleev Labs, that agreement controls to the extent it conflicts with this EULA or the Terms. If you use the Software on behalf of an organization, you represent that you have authority to bind that organization. If you do not have that authority, you may use the Software only for your own permitted individual use or limited organization evaluation. ## 1. Ownership The Software is proprietary and is not open source. Sleev Labs and its licensors retain all rights, title, and interest in the Software, including all intellectual property rights. This EULA grants only the limited rights expressly stated below. All other rights are reserved. ## 2. Individual use Sleev Labs grants one natural person a limited, non-exclusive, non-transferable license to install and use the Software for that person’s own work. Individual use includes personal projects, learning, solo commercial work, freelance work, contracting, consulting, side-business work, and client work. A freelancer or contractor may use the Software internally to create discrete work product for a client. Individual use does not include using the Software to operate, power, perform, or deliver a hosted, managed, automated, outsourced, or recurring service for a client or customer, whether or not the client or customer knows about, accesses, or directly uses Sleev. Such use requires an enterprise subscription or written agreement with Sleev Labs. ## 3. Organization evaluation An organization may conduct one self-serve evaluation of the Software for 30 consecutive calendar days. The evaluation period begins when any person first installs or uses the Software for or on behalf of the organization. During the evaluation period, exactly one evaluator may use the Software on exactly one local machine through either one Sleev account or one unregistered installation, but not both. An unregistered installation is an installation used without a Sleev account. The evaluation period applies to the organization as a whole. Using another employee, contractor, evaluator, account, email address, device, software version, installation, reinstallation, or registered or unregistered mode does not restart or extend it. The evaluation is subject to these additional limits: - Internal evaluation solely to decide whether to purchase Sleev. - Non-production use only, with no reliance on Sleev for regular work or customer deliverables. - No team rollout, shared gateway, CI, automation, central deployment, managed device image, or company-wide setup. - No guaranteed support, uptime, SLA, DPA, or security review unless separately agreed in writing. If an organization wants a longer evaluation, more evaluators, support, a DPA, security review, production use, regular employee use, or any rollout, it must obtain a paid pilot, enterprise subscription, or written agreement from Sleev Labs. ## 4. Organization use Organization use requires a paid pilot, enterprise subscription, or written agreement with Sleev Labs. Organization use includes use where the Software is deployed, managed, coordinated, recommended, required, or relied on by a company, team, institution, client, customer, nonprofit, research group, school, or other organization outside the organization evaluation permitted above. Education, nonprofit, open-source maintainer, and research use do not receive a separate automatic license category. They follow the same individual-use, organization-evaluation, and organization-use rules unless Sleev Labs grants an exception in writing. ## 5. Restrictions Except as expressly allowed by this EULA or by written permission from Sleev Labs, you may not: - Copy, modify, adapt, translate, or create derivative works of the Software. - Resell, sublicense, rent, lease, host, mirror, distribute, bundle, or redistribute the Software. - Make the Software available to others as a hosted service, managed service, shared gateway, service bureau, or outsourced service. - Reverse engineer, decompile, disassemble, or attempt to derive source code from the Software, except to the extent this restriction is prohibited by applicable law. - Bypass or interfere with authentication, licensing, evaluation, billing, usage limits, update integrity, security, or safety controls, or use multiple identities, accounts, machines, or installations to obtain additional evaluation or usage entitlements. - Use the Software in violation of the Terms or applicable law. - Present benchmark, performance, reliability, or competitive claims about Sleev in a false or materially misleading way, or imply that Sleev Labs endorses your results without written permission. ## 6. Benchmarking We welcome good-faith independent benchmarking, including use of Sleev Labs’ published benchmark guidance or suites when available. If you publish results, identify the tested Software version and material configuration and include enough methodology to understand and reproduce them. You may contact Sleev Labs before publication if you want us to validate your methodology or compare it with current benchmark guidance. ## 7. Updates and third-party software The Software may check for, download, install, or prompt you to install updates. Updates may change, add, or remove features. Some versions may stop working or stop receiving support after newer versions are released. Updates remain governed by this EULA unless separate terms accompany them. The Software may include or be distributed with third-party software. Third-party software is governed by its own license terms and notices. This EULA does not limit rights you may have under third-party licenses. ## 8. Termination The organization-evaluation license ends automatically when its 30-day period expires or its limits are exceeded. You may terminate any license by stopping use and uninstalling the Software. Sleev Labs may suspend or terminate your license for a material breach of this EULA, illegal or abusive use, a security risk, or use outside the applicable license scope. When a license ends, you must stop using the Software and delete proprietary copies under your control unless a written agreement says otherwise. Sections that by their nature should continue after termination remain in effect, including ownership, restrictions, disclaimers, liability limits, governing law, and general terms. ## 9. Disclaimers To the fullest extent permitted by law, the Software is provided “as is” and “as available,” without warranties of any kind. Sleev Labs disclaims implied and statutory warranties, including warranties of merchantability, fitness for a particular purpose, title, non-infringement, availability, accuracy, and uninterrupted or error-free operation. Enterprise and pilot customers receive only the warranties stated in their written agreement or order form. ## 10. Liability To the fullest extent permitted by law, Sleev Labs will not be liable for indirect, incidental, special, consequential, exemplary, or punitive damages, or for lost profits, lost revenue, lost data, business interruption, or replacement services. For free self-serve and paid individual self-serve use, Sleev Labs’ total aggregate liability arising out of or relating to the Software, this EULA, or the Terms is limited to the greater of USD $100 or the fees you paid to Sleev Labs during the six months immediately before the event giving rise to the claim. This cap applies across all claims and legal theories and is not cumulative between this EULA and the Terms. For enterprise and pilot customers, liability limits are defined by the applicable written agreement or order form. Nothing in this EULA limits liability that cannot legally be limited. ## 11. Changes to this EULA We may change the Software or this EULA as the product evolves. For users with a Sleev account, we will give notice of material legal changes by email, website notice, or another appropriate product notice. Those users may be asked to expressly accept an updated EULA or may accept it through continued use after the stated effective date. For users without a Sleev account, we provide notice by posting the revised EULA with a stated effective date. By continuing to use the Software on or after that date, you agree to the revised EULA. If you do not agree, you must stop using and uninstall the Software. Minor edits, formatting changes, clarifications, and other non-material changes do not require renewed acceptance. ## 12. General terms This EULA is governed by Delaware law, without regard to conflict-of-law rules. The courts located in Delaware have exclusive jurisdiction and venue for disputes relating to this EULA or the Software, except where applicable consumer law gives you the right to bring a dispute elsewhere or an applicable written agreement specifies different terms. You may not assign this EULA without our written permission. We may assign it in connection with a merger, acquisition, reorganization, or sale of assets. If any provision of this EULA is found unenforceable, the remaining provisions will remain in effect. If we do not enforce a provision, that does not waive our right to do so later. This EULA, the Terms, and any applicable plan or ordering terms are the entire agreement between you and Sleev Labs for self-serve use of the Software, except for any other agreement expressly entered into with Sleev Labs. Headings are for convenience only. For questions or legal notices, email support@sleev.ai. --- ## Data Processing Addendum (DPA) | Sleev URL: https://sleev.ai/dpa Description: Terms governing the processing of personal data by Sleev Labs on behalf of enterprise customers under GDPR, UK GDPR, and CCPA/CPRA. # Data Processing Addendum Effective: September 24, 2026 Effective date: 2026-09-24 This Data Processing Addendum (“DPA”) describes the obligations of Sleev Labs Inc. (“Sleev”) when processing personal data on Customer’s behalf. It forms part of the agreement governing Customer’s use of Sleev (the “Agreement”) only when an order form or other written agreement executed by authorized representatives of Sleev and Customer, including electronically, expressly incorporates this DPA. Publication of, or a link to, this DPA does not by itself make it binding. ## 1. Scope & Roles 1.1 Services. The Sleev gateway runs on Customer-controlled infrastructure and connects to configured upstream endpoints. Sleev also provides hosted account, organization, licensing, and operational services. Details of processing covered by this DPA are set out in Schedule 1. 1.2 Roles. Sleev processes Customer Personal Data on Customer’s behalf and follows Customer’s instructions. Customer must be authorized to give those instructions. Where California privacy law applies, Sleev acts as a service provider or contractor. 1.3 Separate Controller Activities. Sleev acts as an independent controller where it determines the purposes and means of processing for its own customer relationship, invoicing, legal compliance, or protection against fraud and abuse, as described in the [Privacy Policy](/privacy). This distinction depends on the actual processing and does not authorize Sleev to repurpose Customer Personal Data contrary to this DPA. 1.4 Precedence. This DPA prevails over conflicting terms in the Agreement concerning Customer Personal Data. Applicable SCCs and the UK Addendum prevail over conflicting terms in this DPA or the Agreement in accordance with their own precedence provisions. ## 2. Definitions Capitalized terms used but not defined in this DPA have the meanings given in the Agreement or under Applicable Data Protection Law: - “Applicable Data Protection Law” means privacy and data protection laws applicable to the processing of Customer Personal Data under the Agreement, including, where applicable, the EU General Data Protection Regulation (“GDPR”), UK GDPR and Data Protection Act 2018, Swiss Federal Act on Data Protection (“FADP”), and California Consumer Privacy Act as amended by the California Privacy Rights Act (“CCPA/CPRA”). - “Customer Personal Data” means personal data or personal information processed by Sleev on behalf of Customer in providing the services, including data supplied by Customer or collected or generated for Customer through those services. - “Data Subject”, “Controller”, “Processor”, and “Processing” have the meanings given in the GDPR or equivalent concepts under Applicable Data Protection Law. - “Personal Data Breach” means a breach of security leading to the accidental or unlawful destruction, loss, alteration, unauthorized disclosure of, or access to Customer Personal Data transmitted, stored, or otherwise processed by Sleev. - “Standard Contractual Clauses” or “SCCs” means the clauses approved by the European Commission in Implementing Decision (EU) 2021/914 of 4 June 2021 for transfers of personal data to third countries. - “Subprocessor” means a third party engaged by Sleev to process Customer Personal Data on Customer’s behalf. ## 3. Processing Obligations & Instructions 3.1 Documented Instructions. Sleev shall process Customer Personal Data only on Customer’s documented lawful instructions, including instructions concerning international transfers, unless required by law in a manner permitted by Applicable Data Protection Law. The Agreement, this DPA, and Customer’s use of the agreed service features set out those instructions. Customer may provide additional documented lawful instructions consistent with the agreed services. Sleev shall inform Customer before processing required by law unless that law prohibits notification, and shall immediately inform Customer if, in its opinion, an instruction infringes Applicable Data Protection Law. 3.2 Purpose Restrictions. Sleev shall not: - “Sell” or “share” Customer Personal Data (as those terms are defined under the CCPA/CPRA); - Retain, use, or disclose Customer Personal Data for any purpose other than the specific business purposes of performing the services specified in the Agreement; - Retain, use, or disclose Customer Personal Data outside the direct business relationship between Sleev and Customer; or - Combine Customer Personal Data with personal data received from other parties or collected from Sleev’s own interactions with individuals, except as permitted by Applicable Data Protection Law. 3.3 No AI Model Training. Sleev shall not use Customer Personal Data, prompt contents, source code, or operational telemetry to train, retrain, fine-tune, or improve machine learning or artificial intelligence models. 3.4 California Requirements. Where the CCPA/CPRA applies, Sleev shall comply with applicable obligations and provide the same level of privacy protection required by that law. Sleev shall notify Customer if it determines it can no longer meet those obligations. Customer may take reasonable and appropriate steps to verify compliant use of Customer Personal Data and, upon notice, to stop and remediate unauthorized use. 3.5 Customer Responsibilities. Customer is responsible for the lawfulness of its instructions and for providing required notices and obtaining any permissions necessary for the processing it instructs. ## 4. Confidentiality & Personnel 4.1 Confidentiality. Sleev shall ensure that employees, contractors, and agents authorized to process Customer Personal Data are bound by contractual or statutory obligations of confidentiality. 4.2 Access Control. Sleev shall take reasonable steps to ensure the reliability of personnel who process Customer Personal Data and limit access to individuals who require it to perform their obligations under the Agreement. ## 5. Security of Processing 5.1 Security Measures. Sleev shall implement and maintain appropriate technical and organizational measures to protect Customer Personal Data against unauthorized or unlawful processing and accidental loss, destruction, damage, alteration, or disclosure, as detailed in Schedule 2. 5.2 Review & Updates. Sleev shall regularly assess the effectiveness of its security measures as appropriate to the processing risks. Updates shall not materially reduce the overall protection of Customer Personal Data. ## 6. Subprocessors 6.1 Authorization. Customer provides general written authorization for Sleev to engage Subprocessors to provide the agreed services. Subprocessors are identified in the [Subprocessors Directory](/subprocessors). 6.2 Subprocessor Obligations. Sleev shall engage only Subprocessors providing sufficient guarantees of appropriate data protection measures and shall impose the same applicable data protection obligations by written contract for the processing entrusted to them. Sleev remains fully liable to Customer for the performance of each Subprocessor’s obligations. 6.3 Change Notices. Before a new or replacement Subprocessor begins processing Customer Personal Data, Sleev shall inform Customer in writing using the notice contact or method specified in the Agreement. Sleev shall also update the Subprocessors Directory. The notice shall identify the intended change and give Customer a reasonable opportunity to object before the Subprocessor begins processing. 6.4 Objections. Customer may object in writing to an intended Subprocessor change on reasonable data protection grounds before the change takes effect. The parties shall work in good faith to resolve the concern. If they cannot reach a mutually agreeable resolution, either party may terminate the affected hosted services in accordance with the Agreement. 6.5 Customer-Directed Providers. An upstream provider or endpoint engaged by Customer under Customer’s own account or agreement is not a Subprocessor merely because the local gateway connects to it. Where a provider instead processes Customer Personal Data on Sleev’s behalf, this Section 6 applies. ## 7. Personal Data Breaches 7.1 Notification. Sleev shall notify Customer in writing without undue delay after becoming aware of a Personal Data Breach. Notification shall not be delayed until the investigation is complete. 7.2 Details. To the extent known, the notification shall describe: - The nature of the breach, including categories and approximate numbers of affected Data Subjects and records; - A contact for further information; - Likely consequences; and - Measures taken or proposed to address the breach and mitigate its effects. 7.3 Cooperation & Remediation. Sleev shall take immediate, commercially reasonable steps to contain, mitigate, and remediate the breach and assist Customer with its statutory notification obligations. Information unavailable at the initial notification shall be provided in stages without undue further delay. ## 8. Data Subject Requests & Regulatory Assistance 8.1 Data Subject Requests. Taking into account the nature of the processing, Sleev shall assist Customer through appropriate technical and organizational measures, insofar as possible, in responding to Data Subjects exercising their rights under Applicable Data Protection Law. 8.2 Requests Received by Sleev. If an individual asks Sleev to exercise their privacy rights concerning Customer Personal Data, Sleev shall promptly notify Customer and respond only as Customer authorizes or the law requires. 8.3 Regulatory Assistance. Taking into account the nature of the processing and the information available to Sleev, Sleev shall assist Customer in meeting its obligations concerning security, breach notifications, data protection impact assessments, and prior consultation with supervisory authorities. ## 9. Audits 9.1 Information. Sleev shall make available information reasonably necessary to demonstrate compliance with this DPA and applicable processor obligations under Applicable Data Protection Law. 9.2 Audit Rights. Sleev shall allow and contribute to audits and inspections by Customer or its appointed independent auditor, subject to appropriate confidentiality protections. The parties shall use available documentation where it adequately addresses the review. Routine customer audits shall: - Occur no more than once per twelve (12) months during normal business hours; - Be requested with at least thirty (30) days’ advance written notice; - Avoid unreasonable disruption to Sleev’s operations; and - Be conducted at Customer’s sole expense. These routine conditions shall not prevent or delay audits required by law, a competent supervisory authority, or applicable SCCs, including where there are indications of non-compliance. In those circumstances, notice and scope shall be appropriate to the circumstances and mandatory requirements shall prevail. ## 10. Return & Deletion After the provision of services relating to processing ends, Sleev shall, at Customer’s choice, return or delete Customer Personal Data and delete existing copies, unless applicable law requires storage. Any legally required retention is limited to that purpose and lasts only as long as required. Diagnostic uploads remain subject to the seven-day retention period in Schedule 1. ## 11. International Transfers 11.1 When This Applies. These provisions apply when European Economic Area (“EEA”), UK, or Swiss law requires safeguards for an international transfer of Customer Personal Data and no applicable adequacy decision covers it. Before the transfer, the parties shall complete the details in Schedule 1 and assess any additional safeguards required by Applicable Data Protection Law. 11.2 EU SCCs: - The [EU SCCs in Decision (EU) 2021/914](https://eur-lex.europa.eu/eli/dec_impl/2021/914/oj) are incorporated by reference: Module Two where Customer is a controller, and Module Three where Customer is a processor. Customer is the data exporter and Sleev is the data importer. - Options: Clause 7 does not apply. Clause 9(a), Option 2 applies with the advance notice period specified in the Agreement or written transfer supplement. The applicable period must be specified before the restricted transfer begins. The optional language in Clause 11 is omitted. - Law & Courts: Irish law and courts apply under Clauses 17 and 18(b). - Transfer Details: Annex I uses Schedule 1 and the Agreement or written transfer supplement. The supervisory authority is determined under Clause 13. Annex II uses Schedule 2 and any additional agreed security measures. 11.3 UK Addendum: - For UK transfers covered by Section 11.1, the [ICO UK Addendum](https://ico.org.uk/media2/migrated/4019539/international-data-transfer-addendum.pdf) applies to the SCCs in Section 11.2. - Table 1 uses the parties, contacts, start date, and acceptance in the Agreement or transfer supplement. Tables 2 and 3 use the SCC selections and annex information in Section 11.2, respectively. Table 4 permits both parties to exercise the rights in Section 19 of the UK Addendum. - The following is incorporated by reference: Part 2: Mandatory Clauses of the Approved Addendum, being the template Addendum B.1.0 issued by the ICO and laid before Parliament in accordance with s119A of the Data Protection Act 2018 on 2 February 2022, as it is revised under Section 18 of those Mandatory Clauses. 11.4 Swiss Transfers. Where Swiss FADP applies, GDPR references include that law, and the Swiss Federal Data Protection and Information Commissioner supervises the Swiss-law aspects of the transfer. References to EU Member States include Switzerland for Data Subjects’ rights to bring claims where they habitually reside. Any separately applicable GDPR protections and supervisory authority remain unaffected. ## Schedule 1: Details of Processing ### A. Parties & Roles - Customer: The entity identified in the Agreement, acting as controller or processor for the relevant data. - Sleev Labs Inc.: Service provider acting as processor or subprocessor for Customer Personal Data. - Transfer Details: Where Section 11 applies, the Agreement or a written transfer supplement shall identify both parties’ legal names, addresses, contact persons and contact details, relevant activities and roles, acceptance dates and signatures (which may be electronic), processing locations, and the competent supervisory authority determined under SCC Clause 13. These details shall be completed before the restricted transfer begins. Sleev’s privacy contact is support@sleev.ai. ### B. Categories of Data Subjects - Customer’s authorized users, employees, contractors, developers, and team administrators. - Individuals whose personal data is included in support communications or diagnostic material processed under this DPA. ### C. Types of Personal Data - Identity & Contact Data: Names, business email addresses, user identifiers. - Account & Organization Data: Organization names, assigned roles, memberships, and installation or license identifiers. - Usage & Operational Data: Account-linked usage, technical, and security data generated through the services. - Support & Diagnostic Data: Personal data in support communications or diagnostic material voluntarily submitted by Customer’s users. ### D. Special Categories of Data The services do not require special-category data. Customer shall agree appropriate safeguards with Sleev before instructing such processing. If such data is incidentally included in submitted material, it remains protected under this DPA, with access and use limited to the agreed purpose. ### E. Purpose, Nature & Frequency - Collection, transmission, storage, retrieval, analysis, and deletion as needed to provide Customer’s account access, organization and license administration, usage accounting, service security, and technical support under the Agreement. - Processing and transfers occur on an ongoing basis during use of the services; support and diagnostic submissions are voluntarily initiated by Customer’s users. Subprocessors process data for their listed service functions for as long as needed to provide them, subject to the retention limits below. ### F. Retention Customer Personal Data is retained as needed to provide the agreed services during the Agreement and is subject to deletion or return under Section 10. Diagnostic uploads are retained for no longer than seven (7) days. ## Schedule 2: Technical & Organizational Measures The following measures apply to the services. Additional customer-specific measures shall be recorded in the Agreement where applicable. - Deployment Boundary: The gateway runs on Customer-controlled infrastructure and defaults to a loopback listen address. Customer manages its endpoint security, local storage protection, and any network-reachable deployment. - Encryption: The hosted control-plane API uses HTTPS. Managed Google Cloud SQL databases and Google Cloud Storage use encryption at rest. Registered remote model-provider endpoints use HTTPS/WSS; transport and access protection for custom endpoints are Customer-managed. - Service Access: Access to customer account and organization data requires authentication and authorization. Organization access is checked against membership and role permissions. Personnel access is subject to the confidentiality and access restrictions in Section 4. - Software Checks: Development workflows include automated linting, type checking, and tests. Sleev shall assess and address security vulnerabilities in the services as appropriate to the risk. - Incident Handling: Security reports may be sent to support@sleev.ai. Sleev shall investigate Personal Data Breaches and provide notification and assistance under Sections 7 and 8. - Data Lifecycle: Diagnostic uploads are voluntarily initiated by users. Managed diagnostic logging includes content-redaction measures, and periodic automated cleanup is configured. The retention and deletion obligations in Schedule 1 and Section 10 apply. ## Contact Sleev Labs Inc. Email: support@sleev.ai --- ## Subprocessors Directory | Sleev URL: https://sleev.ai/subprocessors Description: Third-party service providers engaged by Sleev Labs Inc. to process personal data in connection with Sleev hosted services. # Subprocessors Directory Last updated: September 24, 2026 Effective date: 2026-09-24 To support the delivery of Sleev’s hosted control-plane services, account management, organization administration, and subscription billing, Sleev Labs Inc. (“Sleev”) engages third-party service providers (“Subprocessors”) that may process Customer Personal Data. This page describes service providers supporting Sleev’s hosted services and the subprocessor notice process under our [Data Processing Addendum (DPA)](/dpa). A provider’s role depends on the processing it performs under the applicable agreement. ## 1. Local Gateway vs. Hosted Services As detailed in our [Privacy Policy](/privacy), the Sleev software operates as a local-first gateway residing on the customer’s machine (127.0.0.1). - Local Gateway: The gateway runs on customer-controlled infrastructure and connects to configured upstream endpoints. - Hosted Services: Service providers support account administration, licensing, billing, operational services, and optional diagnostic uploads. Data collection is described in the [Privacy Policy](/privacy). ## 2. Infrastructure & Service Subprocessors The following third-party entities are currently engaged by Sleev Labs Inc. to process personal data: Provider Service Google Cloud Platform Hosted API infrastructure, managed database, and diagnostic storage WorkOS Account authentication and organization identity management Stripe Payment processing and subscription billing Resend Transactional email delivery For transfers covered by DPA Section 11, the parties shall document the applicable processing locations and transfer safeguards before the transfer begins. ## 3. Direct Upstream AI Model Providers An upstream provider engaged under the customer’s own account or agreement is not a Sleev Subprocessor merely because the local gateway connects to it. Where a provider processes Customer Personal Data on Sleev’s behalf, DPA Section 6 applies. ## 4. Data Protection Requirements DPA Section 6 requires appropriate data protection guarantees and written terms for Subprocessors. International transfers must meet the requirements in DPA Section 11. ## 5. Notification & Objection Process Sleev provides customers with advance notice of any planned addition or replacement of a Subprocessor: - Advance Notice: Before an intended new or replacement Subprocessor begins processing Customer Personal Data, Sleev will notify each Customer with an effective DPA and update this directory. The Agreement or transfer supplement may specify a notice period. - Notification Method: Direct notice will use the notice contact or method specified in the applicable Agreement. - Objection Procedure: In accordance with Section 6.4 of the [DPA](/dpa), Customer may object in writing on reasonable data protection grounds before the intended change takes effect. ## 6. Questions & Inquiries If you have questions regarding our Subprocessors, compliance, or third-party risk management practices, please contact: Sleev Labs Inc. Email: support@sleev.ai --- ## Security | Sleev URL: https://sleev.ai/security Description: How Sleev keeps your code local, protects account data, and handles security reports. # Security Last updated: October 11, 2026 Sleev is built so your code never has to leave your machine. This page summarizes what Sleev can and can’t see and how we protect the data we do handle. Our [Privacy Policy](/privacy) and [Data Processing Addendum (DPA)](/dpa) describe our commitments in full. ## What Sleev sees - Account and billing details, such as your email, name, organization, and plan. - Usage metadata, such as request counts, token totals, models, and software versions. - Diagnostic logs, only when you choose to upload them. ## What Sleev never sees - Your prompts, code, or file contents. - Conversation history, model responses, or tool inputs and outputs. - Your provider API keys. - Your card details. Stripe processes payments. The Sleev gateway runs on your device and listens on a loopback address (127.0.0.1) by default. It sends your requests directly to the AI provider or endpoint you configure, using your provider credentials locally. If you make the gateway reachable on a network, securing that network is up to you. ## Hosted services Sleev’s hosted services handle accounts, organizations, licensing, and billing. - Encryption: The hosted API is served over HTTPS. Our managed databases and storage on Google Cloud are encrypted at rest. - Access control: Account and organization data require authentication. Organization access is checked against membership and role. - Internal access: Sleev’s internal systems require two-factor authentication. - Providers: The service providers we use are listed in our [Subprocessors Directory](/subprocessors). ## Diagnostics Diagnostic logs exclude prompts, code, conversation history, model responses, tool content, and provider API keys. They are deleted within seven days. ## Network connections Besides the AI providers you configure, Sleev connects to these domains. If you’re behind a corporate proxy or firewall, allow them. Domain Used for sleev.ai Browser sign-in and the installer api.sleev.ai Account, licensing, and usage sync api.workos.com Sign-in sessions storage.googleapis.com/sleeve-releases Software updates models.dev Model pricing for local usage analytics Machines that can’t reach these domains can use an [offline license](/docs/offline-licensing). ## Reporting a vulnerability If you believe you’ve found a security issue in Sleev, email support@sleev.ai with the details and steps to reproduce it. Please give us a chance to fix the issue before disclosing it publicly. Our [security.txt](/.well-known/security.txt) lists the same contact. --- ## Contact Us | Sleev URL: https://sleev.ai/contact Description: How to reach the Sleev team for support, feedback, business, privacy, and terms questions. # Contact us Pick the channel that fits and a real person will get back to you. [Join the Sleev Discord](https://discord.gg/W36rxJswbG) Support and legal notices Help with installing the gateway, configuring your harness or coding tool, privacy questions, terms questions, legal notices, or anything that is not working the way it should. support@sleev.ai General Pilot scope, purchasing, security review, partnerships, and anything else that is not a support or legal notice issue. contact@sleev.ai