# selan.ai sitemap Every public page on selan.ai, grouped by what it is for. Each page has a Markdown twin, linked after its line, and the page's own URL answers `Accept: text/markdown` with the same file. The same pages, dated, are in [sitemap.xml](https://selan.ai/sitemap.xml). ## Product What Selan is, how it works, and who makes it: `/`, `/about`, `/browser`, `/contact`, `/en/modernise`, `/en/modernise/uk`, `/how-it-works`, `/lt/modernise`, `/slack`, `/status`. - [See and control what your team spends on AI · Selan](https://selan.ai/): Selan shows who spends what across every AI account and model, caps what they can spend, and stops secrets before they reach the provider. Markdown: - [About Selan](https://selan.ai/about): Who makes Selan: Selan UAB, registered in Lithuania, building one company setup for the AI accounts a team already pays for. How to reach us, and more. Markdown: - [Selan Browser Extension · your agent, in your own Chrome · Selan](https://selan.ai/browser): Claude Code and Claude Desktop open tabs in the Chrome you already use, with the sessions you already have. Only in their own tab group, and only on the sites you allow. Free, no account. Markdown: - [Talk to us · Selan](https://selan.ai/contact): Tell us what your team runs and where to reach you. A person reads every message and replies to the address you leave. Markdown: - [Selan Web · convert your old website into one ready for the age of AI agents](https://selan.ai/en/modernise): We modernise your website and AI agents keep it maintained: checks, fixes and every change you ask for. One monthly price with credits for more agents. Markdown: - [Selan Web UK · convert your old website into one ready for the age of AI agents](https://selan.ai/en/modernise/uk): We modernise your company's website and AI agents keep it maintained: checks, fixes and every change you ask for. £39 a month, no setup fee, leave any time. Markdown: - [How an AI gateway works for Claude Code and Codex · Selan](https://selan.ai/how-it-works): Add your accounts and keys, give your team one CLI, and Selan routes each Claude Code and Codex request to the right account, with spend limits per user. Markdown: - [Selan Web · seną svetainę paverčiame nauja, paruošta AI agentų erai](https://selan.ai/lt/modernise): Atnaujiname jūsų svetainę, o AI agentai ją prižiūri: tikrina, taiso ir atlieka kiekvieną jūsų prašomą pakeitimą. Viena mėnesio kaina su kreditais agentų darbui. Markdown: - [Selan for Slack — AI agents that post under their own names](https://selan.ai/slack): Selan agents read the Slack channels you invite them to, post under their own names and answer when tagged. Every post says who asked and links the run. Markdown: - [Status · the AI providers Selan uses, and Selan](https://selan.ai/status): Whether Claude, OpenAI, Vertex AI, OpenRouter and Kimi are answering right now, measured from Selan's own traffic beside what each provider says about itself, and Selan's own status. Updated every minute. Markdown: ## Reactor Selan's second product, in early access, whose micro agents maintain and build your system while you sleep: `/reactor`, `/reactor/`. - [Reactor · micro agents that maintain and build your system while you sleep · Selan](https://selan.ai/reactor): Micro agents that maintain and build your system while you sleep. Each does one job overnight and leaves you a merge request to review in the morning. Markdown: - [Agents for Selan Reactor · Selan](https://selan.ai/reactor/agents): Remote agents that already know the job, one Markdown file each on GitHub. Paste one line into Claude Code and it sets the agent up in your Reactor. Markdown: - [AGENTS.md Updater · Reactor agents · Selan](https://selan.ai/reactor/agents/agents-md-updater): Keeps a repository's AGENTS.md true and short by Selan's practices, with commands that work, traps already hit, and nothing the code already says. Markdown: - [CSS Cleaner · Reactor agents · Selan](https://selan.ai/reactor/agents/css-cleaner): Removes unused and duplicated CSS and folds repeated styles into shared components and design tokens, in React, Vue, Svelte, Angular or plain CSS. Markdown: - [Dead Code Cleaner · Reactor agents · Selan](https://selan.ai/reactor/agents/dead-code-cleaner): Removes code and dependencies nothing uses, one small merge request a week, each removal proven unused. Markdown: - [Dependency Updater · Reactor agents · Selan](https://selan.ai/reactor/agents/dependency-updater): Updates npm, yarn or pnpm dependencies only where it is safe, known vulnerabilities first, never a release too new to trust. Markdown: - [E2E Test Writer · Reactor agents · Selan](https://selan.ai/reactor/agents/e2e-test-writer): Plans end-to-end coverage of a whole repository and works toward 100%, a few endpoints or flows a run, each test driving the real running service and its database, one merge request a run. Markdown: - [Flake Hunter · Reactor agents · Selan](https://selan.ai/reactor/agents/flake-hunter): Finds the flakiest test in a repository's CI history and fixes why it is flaky, never with a retry. Markdown: - [LLM Visibility Agent · Reactor agents · Selan](https://selan.ai/reactor/agents/llm-visibility): Asks ChatGPT, Claude, Gemini and Perplexity the same buyer questions every week, records whether they name and cite your company, and opens one merge request at a time that makes your website easier for them to find and quote. Markdown: - [MR Watcher · Reactor agents · Selan](https://selan.ai/reactor/agents/mr-watcher): Keeps a repository's open merge requests moving, nudging the ones that wait on a person and fixing mechanical conflicts. Markdown: - [Nightly Beta · Reactor agents · Selan](https://selan.ai/reactor/agents/nightly-beta): Every morning gives testers last night's build of a mobile app on TestFlight and a Google Play testing track, with notes they understand, and sets up the nightly build in CI when it is missing. Markdown: - [Optimistic UI Engineer · Reactor agents · Selan](https://selan.ai/reactor/agents/optimistic-ui): Finds the buttons, toggles and forms where people wait on the server before seeing anything, and makes them answer at once: optimistic updates with rollback where that is safe, an honest pending state where it is not, each proven by a test, one merge request a run. Markdown: - [Package Publisher · Reactor agents · Selan](https://selan.ai/reactor/agents/package-publisher): Publishes a new version of a package when its default branch has changes users would notice, to npm, PyPI, crates.io or its own registry, then tags it and writes the release. Markdown: - [Prod Deployer · Reactor agents · Selan](https://selan.ai/reactor/agents/prod-deployer): Deploys a repository's default branch to production when its pipeline, merged changes and staging logs are clean and nothing is being reverted, then watches production's logs and rolls back a deploy that raises errors. Markdown: - [Security Reviewer · Reactor agents · Selan](https://selan.ai/reactor/agents/security-reviewer): Keeps working out how to make one repository more secure, area by area, and fixes the real holes it finds, each with an attacker, a path at file:line and a test that fails without the fix. Markdown: - [Sentry Fixer · Reactor agents · Selan](https://selan.ai/reactor/agents/sentry-fixer): Drives the unresolved Sentry issues of one repository to zero, one merge request with a test per issue. Markdown: - [SEO Improver · Reactor agents · Selan](https://selan.ai/reactor/agents/seo-improver): Grows a website's clicks from search, one merge request a run: sharpens the page where search data shows demand or writes a fully sourced new one, fixes one technical SEO issue, and measures every change it made. Markdown: - [Store Releaser · Reactor agents · Selan](https://selan.ai/reactor/agents/store-releaser): Ships a mobile app to the App Store and Google Play by itself, rolls each version out over 7 days, halts a version that raises new errors and ships its own hotfix. Markdown: - [Unit Test Writer · Reactor agents · Selan](https://selan.ai/reactor/agents/unit-test-writer): Gives what was merged since its last run unit tests that fail if the change is reverted, in the repository's own framework, one merge request a run. Markdown: - [Our mission · Reactor · Selan](https://selan.ai/reactor/mission): Why Reactor is lots of small agents with one job each, what every Reactor feature has to keep true, and what is built so far. Markdown: ## Pricing The plans, and what counts as a seat: `/pricing`. - [Pricing · Selan](https://selan.ai/pricing): Flat monthly plans sized by seats: free for up to 3, then bands up to 50. A seat is a person in your workspace or a CI token; a Reactor agent takes none. Your provider accounts keep billing you directly. Markdown: ## Trust What crosses Selan and what is kept, and the legal pages: `/trust`, `/privacy`, `/privacy/`, `/terms`. - [Privacy Policy — Selan](https://selan.ai/privacy): What personal data Selan holds and what it does not, the few cases where part of a reply or a run is kept, what the Selan Reactor phone apps send, where it lives, who processes it, for how long, and how to delete your account. Markdown: - [Selan MCP Extension Privacy Policy — Selan](https://selan.ai/privacy/extension): What the Selan MCP Extension for Chrome reads, where it sends it, how long anything is kept, and what it never touches: cookies, saved passwords, browsing history, and every tab it did not open. Markdown: - [Selan Slack App Privacy Policy — Selan](https://selan.ai/privacy/slack): What the Selan Slack app stores, how it is encrypted, who can reach the token, how long anything is kept, and how disconnecting or uninstalling removes it. Markdown: - [Terms of Service — Selan](https://selan.ai/terms): The terms on which Selan provides its service: what Selan does, what it does not do, and where responsibility for your provider accounts and their usage sits. Markdown: - [Trust & data · what crosses Selan, and what we keep](https://selan.ai/trust): The proxy receives the full request body. It streams through, and no request body is logged or stored. What we keep is metadata about the call, plus what a failed reply leaves in our logs, named on the page. Markdown: ## Solutions One page per problem a team brings to Selan: `/solutions/`. - [Claude Code usage analytics — per repo and person · Selan](https://selan.ai/solutions/insights): Which repository the hours went to, which people were in it, and how much of the context came back out of the cache instead of being paid for twice. Markdown: - [Any model in Claude Code, on your own accounts · Selan](https://selan.ai/solutions/models): Connect the AI accounts your company already pays for, give the team one sign-in, and let people use the models you enable. Markdown: - [Keep secrets out of Claude Code — redacted in flight · Selan](https://selan.ai/solutions/secrets): With secret scanning on, Selan replaces each credential Gitleaks' rules find in a Claude Code, Codex or Claude Desktop request before the provider gets it. Markdown: - [Claude Code spend per developer — see it, cap it · Selan](https://selan.ai/solutions/spend): See what each developer spends in Claude Code, Codex and Claude Desktop on the AI accounts you already pay for, and cap each person by day, week or month. Markdown: ## Docs Installing, running and administering Selan; `/docs` is the hub: `/docs`, `/docs/`, `/docs//`. - [Claude Code, Codex & Desktop on company accounts · Selan](https://selan.ai/docs): Install selan, sign in once, and Claude Code, Codex and Claude Desktop run on your company's accounts, with your usage and limits on the statusline. Markdown: - [Claude Code agents in GitHub Actions & GitLab CI · Selan](https://selan.ai/docs/agents): selan agent-run runs a Claude Code agent with nobody watching: an agent file and an API token in a GitHub Actions or GitLab CI job, under company limits. Markdown: - [Claude Code code review agent for GitHub & GitLab · Selan](https://selan.ai/docs/agents/code-review): A Claude Code agent reviews every pull request, with a severity table and a fix per finding: one comment on GitHub, a Code Quality report in GitLab CI. Markdown: - [Claude Code writes too many comments? Remove them · Selan](https://selan.ai/docs/agents/minimise-comments): An agent that deletes comments that restate the code and keeps the ones that explain why. It runs on a small, fast model, from a pre-commit hook or in CI. Markdown: - [Messages API gateway for Claude, GPT and DeepSeek · Selan](https://selan.ai/docs/api): Point an Anthropic SDK's base URL at Selan's gateway and send one token: your scripts run on your company's accounts, metered in Usage and Logs per token. Markdown: - [Agent apps on company accounts — selan app · Selan](https://selan.ai/docs/apps): Run Hermes Agent, Paperclip and other agent apps on your company's Selan accounts with selan app or two lines of bash, from public scripts you read first. Markdown: - [Let your agent use your Chrome — Browser Extension · Selan](https://selan.ai/docs/browser-extension): Let Claude Code, Claude Desktop, ChatGPT, Codex, Cursor or any MCP client use your own Chrome, kept to one tab group and the sites you allow. No account. Markdown: - [Claude Agent SDK — run it on company accounts · Selan](https://selan.ai/docs/claude-agent-sdk): Run the Claude Agent SDK in TypeScript or Python on your company's accounts: install it, set two environment variables, write an agent, add tools over MCP. Markdown: - [Claude Desktop on a company account — setup · Selan](https://selan.ai/docs/claude-desktop): Run Claude Desktop on your company's account in seven steps: import one settings file, relaunch, sign in. It shares Claude Code's limits, Usage and Logs. Markdown: - [Import claude.ai chats into Claude Desktop · Selan](https://selan.ai/docs/claude-desktop/import): Move claude.ai chats, projects and attachments to Claude Desktop. An extension builds the archive for Desktop to import; one command adds scheduled tasks. Markdown: - [Codex CLI on your company's accounts — selan codex · Selan](https://selan.ai/docs/codex): Run OpenAI Codex on your company's accounts with selan codex. The company's models are in /model, the limits are the same, and selan usage shows yours. Markdown: - [Codex SDK — run it on your company's accounts · Selan](https://selan.ai/docs/codex-sdk): Run the Codex SDK on your company's accounts: install @openai/codex-sdk, set a base URL and token, start a thread, sandbox the agent, or call Responses. Markdown: - [Hermes Agent on company accounts — selan app hermes · Selan](https://selan.ai/docs/hermes): Run Nous Research's Hermes Agent on your company's Selan accounts: its models, connectors and skills, with selan app, by hand, or on your own server. Markdown: - [Claude Code in JetBrains IDEs on company accounts · Selan](https://selan.ai/docs/jetbrains): One command sets up Claude Code in IntelliJ IDEA, GoLand, WebStorm, PyCharm and Android Studio, so IDE sessions run on your company's accounts. Markdown: - [Selan MCP — usage data and remote agents · Selan](https://selan.ai/docs/mcp): One MCP server lets Claude Code read your company's Selan data, such as spend and usage, and manage Reactor remote agents as the signed-in member, by role. Markdown: - [MCP servers for Claude: GitHub, Jira, Slack and more · Selan](https://selan.ai/docs/mcp-connections): Every MCP server Selan connects Claude to: GitHub, GitLab, Jira, Linear, Notion, Slack, Sentry, Google and more. What each does, and how you sign in. Markdown: - [OpenCode on company accounts — selan opencode · Selan](https://selan.ai/docs/opencode): Run OpenCode on your company's accounts: the company's models, connectors and skills, the same limits, and nothing written to OpenCode's config. Markdown: - [Paperclip on company accounts — selan app paperclip · Selan](https://selan.ai/docs/paperclip): Run Paperclip, a web app for a company of AI agents, on your company's Selan accounts: with selan app, by hand, or on your own server behind an SSH tunnel. Markdown: - [Your AI agents on Android — the Selan Reactor app · Selan](https://selan.ai/docs/phone-app): Selan Reactor for Android: your workspaces, agents and runs on your phone. Answer what agents ask you, chat with the Expert and call Selan by voice. Markdown: - [Reduce Claude Code token usage per repository · Selan](https://selan.ai/docs/practices): selan-practices reads your repository's Claude Code sessions on your machine, shows what they cost, and proposes changes to AGENTS.md, skills and checks. Markdown: - [Claude, GPT, Kimi and DeepSeek on your own accounts · Selan](https://selan.ai/docs/providers): Claude, GPT, Kimi, DeepSeek, Qwen and GLM on your company's own accounts through one endpoint. Subscriptions cover five; Qwen runs per token on OpenRouter. Markdown: - [selan CLI reference — commands, env vars, errors · Selan](https://selan.ai/docs/reference): Every selan command, environment variable and error message, including selan usage and SELAN_TOKEN, and what differs on Windows. Markdown: - [Self-hosted runners for Reactor remote agents · Selan](https://selan.ai/docs/runners): Run Reactor remote agents on a server or Mac your company keeps on, or on a VM per run in your own Google Cloud: add a runner, install it, remove it. Markdown: - [Claude Code secret scanning and redaction · Selan](https://selan.ai/docs/secrets): With secret scanning on, Selan replaces credentials in Claude Code and Desktop requests before the model sees them. What is scanned, and a safe test token. Markdown: - [Claude Code statusline usage — what each meter means · Selan](https://selan.ai/docs/statusline): How to check Claude Code usage on a company account: the statusline shows your company's 5h and 7d headroom and your own budget; selan usage prints both. Markdown: - [Claude on Google Vertex, billed to your own project · Selan](https://selan.ai/docs/vertex): Run Claude on a Google Cloud project you own, billed to your own Google account. What to create, what to grant, and the two steps that catch people out. Markdown: - [Virtual identities on your Google Workspace domain · Selan](https://selan.ai/docs/virtual-identities): Connect your Google Workspace, prepare Cloud Identity Free, then create, reset and delete Google accounts for agents. Owner-only, with its limits. Markdown: ## Guides One question per page, answered for someone who has not used Selan; `/guides` is the hub: `/guides`, `/guides/`. - [Claude Code guides — cost, limits, keys, models · Selan](https://selan.ai/guides): Every Selan guide and comparison: what Claude Code costs a team, its limits, usage per person, keys, EU residency, other models and a policy template. Markdown: - [AI coding policy template for Claude Code and Codex · Selan](https://selan.ai/guides/ai-coding-policy-template): A copy-paste AI coding tool policy for Claude Code and Codex: accounts, keys, secrets, models, spend, agents and leavers, each clause with its control. Markdown: - [Claude Code 529 overloaded error — what to do · Selan](https://selan.ai/guides/claude-code-529-overloaded): Claude Code's 529 overloaded error means Anthropic is at capacity, not that you hit a limit. What to do now, in CI, and why a second seat does not help. Markdown: - [Claude Code cost per developer — a day and a month · Selan](https://selan.ai/guides/claude-code-cost-per-developer): Anthropic puts Claude Code at about $13 per developer per active day and $150–250 a month. Why the spread is wide, and how to measure and cap it. Markdown: - [Claude Code EU data residency and GDPR, with Codex · Selan](https://selan.ai/guides/claude-code-eu-data-residency): Anthropic's API runs Claude globally or in the US; Vertex's EU location keeps it in the EU. What Selan serves Claude Code and Codex, and where calls land. Markdown: - [Claude Code gateway — which kind a team needs · Selan](https://selan.ai/guides/claude-code-gateway): A Claude Code gateway sits between Claude Code and the model provider. The kinds a team can choose, who runs each, and when each one is the right answer. Markdown: - [Claude Code GitHub Actions — team credentials · Selan](https://selan.ai/guides/claude-code-github-actions): How to run Claude Code in GitHub Actions for a team: which credential to store, what a run costs, how to cap it, and a token per job instead of one key. Markdown: - [Claude Code GitLab CI — team setup and credentials · Selan](https://selan.ai/guides/claude-code-gitlab-ci): How to run Claude Code in GitLab CI for a team: which credential to store, why a protected variable can miss an MR, how to cap a run, a token per job. Markdown: - [Claude Code prompt caching — write vs read cost · Selan](https://selan.ai/guides/claude-code-prompt-caching): A cache write costs 12.5 to 50 times a cache read on Claude's current models. What Claude Code caches, what makes it miss, and how a team keeps it warm. Markdown: - [Claude Code rate limit reached — resets and fixes · Selan](https://selan.ai/guides/claude-code-rate-limits): Claude Code stops when your plan fills its 5-hour session limit or its weekly limit. When each one resets, and what each way past it costs. Markdown: - [Claude Code skip permissions safely — team setup · Selan](https://selan.ai/guides/claude-code-skip-permissions): How to run Claude Code with --dangerously-skip-permissions safely: what it turns off, where to run it, safer modes, and limits for a team and CI. Markdown: - [Claude Code spend limit reached — whose cap it is · Selan](https://selan.ai/guides/claude-code-spend-limit-reached): Claude Code stops at a spend limit when a cap someone set on paid usage is full. Which message means whose limit, when it lifts, and one per developer. Markdown: - [Claude Code usage per developer and per repository · Selan](https://selan.ai/guides/claude-code-usage-per-developer): How to see Claude Code usage and cost per developer and per repository: Anthropic's analytics, OpenTelemetry, and a gateway that adds Codex and Desktop. Markdown: - [Claude Code without an API key on every laptop · Selan](https://selan.ai/guides/claude-code-without-api-keys): Hold the Anthropic and OpenAI keys on a gateway and sign each developer in. The only credential left on a laptop is a session token you can revoke. Markdown: - [Codex CLI usage tracking — per developer and team · Selan](https://selan.ai/guides/codex-usage-tracking): Track Codex CLI usage per developer with /status and OpenAI's workspace analytics, see where they stop, and put Codex beside Claude Code per person. Markdown: - [DeepSeek in Claude Code — for a whole team · Selan](https://selan.ai/guides/deepseek-in-claude-code): DeepSeek runs in Claude Code through any endpoint that speaks Anthropic's API. Run it for a team on Ollama Cloud or OpenRouter, with no key on a laptop. Markdown: - [Gemini in Claude Code — on your own Google Cloud · Selan](https://selan.ai/guides/gemini-in-claude-code): Claude Code speaks Anthropic's API, so Gemini needs a gateway that translates. Run Gemini 3.8 Flash on your own Vertex project or OpenRouter. Markdown: - [Gemma 4 in Claude Code — local, or for a whole team · Selan](https://selan.ai/guides/gemma-in-claude-code): Gemma 4 runs in Claude Code through Ollama, locally or on its cloud. Run it for a team on Ollama Cloud or OpenRouter, beside Claude, no key on a laptop. Markdown: - [GLM 5.3 in Claude Code — for a whole team · Selan](https://selan.ai/guides/glm-in-claude-code): GLM 5.3 runs in Claude Code through Z.ai's Anthropic-format endpoint. Run it for a team on OpenRouter or Ollama Cloud, beside Claude, no key on a laptop. Markdown: - [GPT-5 and GPT-6 in Claude Code — for a whole team · Selan](https://selan.ai/guides/gpt-in-claude-code): Claude Code speaks Anthropic's API, not OpenAI's, so GPT needs a gateway that translates. Run it on your ChatGPT plan, an OpenAI key or OpenRouter. Markdown: - [Kimi K3 in Claude Code — for a whole team · Selan](https://selan.ai/guides/kimi-in-claude-code): Kimi K3 runs in Claude Code through any endpoint speaking Anthropic's API. Run it for a team on OpenRouter or Ollama Cloud, with no key on a laptop. Markdown: - [Claude Max plan vs API cost — the break-even · Selan](https://selan.ai/guides/max-plan-vs-api-cost): Max 5× costs $100 a month and Max 20× $200. A seat is cheaper than the API once a developer's traffic would cost more than that at list price. Markdown: - [MiniMax in Claude Code — M3 for a whole team · Selan](https://selan.ai/guides/minimax-in-claude-code): MiniMax runs in Claude Code through its Anthropic-format API. Run M3 for a team on OpenRouter or Ollama Cloud, beside Claude, no key on a laptop. Markdown: - [Ollama in Claude Code — local or cloud, for a team · Selan](https://selan.ai/guides/ollama-in-claude-code): Ollama runs Claude Code on a local or cloud model through its Anthropic-compatible API. Run Ollama's cloud for a team through Selan, no key on a laptop. Markdown: - [OpenCode with a Claude subscription — what works · Selan](https://selan.ai/guides/opencode-claude-subscription): Can OpenCode use a Claude Pro, Max or Team plan? What Anthropic allows, why the plugins are a risk, what works instead, and OpenCode on a team's accounts. Markdown: - [Qwen in Claude Code — Qwen3.8 for a whole team · Selan](https://selan.ai/guides/qwen-in-claude-code): Qwen runs in Claude Code through Alibaba's Anthropic-format endpoints. Run Qwen3.8 Max for a team on OpenRouter, beside Claude, no key on a laptop. Markdown: ## Comparisons Selan beside another tool, including where that tool is the better answer: `/compare/`. - [Bifrost alternative — for a team's coding tools · Selan](https://selan.ai/compare/bifrost): Bifrost is an open-source Go gateway you run for models and MCP tools, reached by virtual key. Selan is hosted, one sign-in per person. When each fits. Markdown: - [ccusage alternative — for a whole team, not a laptop · Selan](https://selan.ai/compare/ccusage): ccusage is free and reads one machine's logs. Selan records a team's Claude Code, Codex and Claude Desktop usage per person. When each one fits. Markdown: - [Claude apps gateway alternative — hosted, with Codex · Selan](https://selan.ai/compare/claude-apps-gateway): Anthropic's Claude apps gateway is free and self-hosted, for Claude on your own cloud. Selan is hosted and adds Codex and subscriptions. When each fits. Markdown: - [Claude Code Router alternative — for a team · Selan](https://selan.ai/compare/claude-code-router): Claude Code Router is a free local gateway that routes Claude Code to the providers you pick. Selan is a hosted gateway a team signs in to. When each fits. Markdown: - [Cloudflare AI Gateway alternative — for coding tools · Selan](https://selan.ai/compare/cloudflare-ai-gateway): Cloudflare AI Gateway is a proxy Claude Code reaches with an account-wide token. Selan is hosted, one sign-in per person. Where each one fits. Markdown: - [Kong AI Gateway alternative — for coding tools · Selan](https://selan.ai/compare/kong-ai-gateway): Kong AI Gateway puts Claude Code behind your identity provider, on a data plane you run. Selan is hosted, one sign-in per person. Where each fits. Markdown: - [LiteLLM alternative — for a team's coding tools · Selan](https://selan.ai/compare/litellm): LiteLLM is an open-source proxy you run for your own apps. Selan is a hosted gateway for a team's Claude Code, Codex and Claude Desktop. When each fits. Markdown: - [MLflow AI Gateway alternative — for coding tools · Selan](https://selan.ai/compare/mlflow-ai-gateway): MLflow AI Gateway traces Claude Code on a server you run, on each developer's Anthropic login. Selan is hosted, one sign-in per person. Where each fits. Markdown: - [nexos.ai alternative — for Claude Code and Codex · Selan](https://selan.ai/compare/nexos): nexos.ai is an AI platform and gateway billed in credits. Selan runs a team's Claude Code and Codex on your subscriptions, and can use your nexos.ai key. Markdown: - [OpenRouter alternative — for a team's coding tools · Selan](https://selan.ai/compare/openrouter): OpenRouter sells 500+ models per token. Selan runs a team's Claude Code and Codex on your subscriptions first, with OpenRouter behind them. When each fits. Markdown: - [Portkey alternative — for a team's coding tools · Selan](https://selan.ai/compare/portkey): Portkey, now Prisma AIRS AI Gateway, routes a company's AI apps and Claude Code on Portkey API keys. Selan is a gateway a team signs in to. When each fits. Markdown: - [TrueFoundry alternative — for coding tools · Selan](https://selan.ai/compare/truefoundry): TrueFoundry's AI Gateway routes Claude Code on an API key or each developer's own seat. Selan is hosted, one sign-in per person. Where each fits. Markdown: - [Vercel AI Gateway alternative — for coding tools · Selan](https://selan.ai/compare/vercel-ai-gateway): Vercel AI Gateway runs Claude Code on a gateway key per developer or their own Claude plan. Selan is one sign-in per person. Where each one fits. Markdown: ## Blog New features and how to use them, and news from Selan, a page per post, newest first: `/blog`, `/blog/`. - [Selan blog: new features and what changed · Selan](https://selan.ai/blog): New features in Selan, how they work and what changed for the teams that use it, newest first. Every post links to the docs that describe Selan today. Markdown: - [The Selan Chrome extension: agents in your own browser · Selan](https://selan.ai/blog/2026-10-15-selan-chrome-extension): Claude Code, Claude Desktop and Selan's remote agents can open and use tabs in your own Chrome, where you are already signed in. Here is how to set it up and use it. Markdown: - [Slack in Selan: the company bot and Slack Personal · Selan](https://selan.ai/blog/2026-10-15-slack-bot-and-personal): Agents post and take replies in Slack as your company's bot, and each developer's Claude reads and writes Slack as them. Here is how each works and how to set it up. Markdown: - [Welcome to the Selan blog · Selan](https://selan.ai/blog/2026-10-15-welcome-to-the-blog): New features in Selan and how to use them, in one place. Every post links to the docs that describe the feature in full. Markdown: - [New connectors: GitHub, Firebase, Search Console and more · Selan](https://selan.ai/blog/2026-10-16-new-connectors): Claude can now open GitHub pull requests with no GitHub token on its machine, work in Firebase, read Search Console and Analytics, and reach Cloudflare, Hostinger and Higgsfield. Markdown: - [Reactor: a fleet of Claude Code agents that runs without you · Selan](https://selan.ai/blog/2026-10-16-reactor): Reactor starts your company's Claude Code agents on a schedule, a webhook, a Slack reply or each other, in the cloud or on your own machines. What it does today, and where it is going. Markdown: - [Selan practices: find what Claude Code spends, and cut it · Selan](https://selan.ai/blog/2026-10-16-selan-practices): A long, stale CLAUDE.md is paid for on every request, and missing checks let broken changes through. Selan practices reads your sessions on your machine and proposes the fixes, with the numbers behind each. Markdown: - [Top Claude Code skills, ranked from GitHub every day · Selan](https://selan.ai/blog/2026-10-16-top-claude-code-skills): A daily list of the 15 most-starred Claude Code skills on GitHub: which are moving this week and this month, what each does and how to install it. Markdown: - [whatileaked: find the keys your coding agent already sent · Selan](https://selan.ai/blog/2026-10-16-whatileaked): A free, MIT-licensed command that scans your Claude Code, Codex, Cursor and opencode history for credentials already sent to a model provider, and cleans them up. No network calls, no dependencies. Markdown: - [One laptop, several companies: accounts in the Selan CLI · Selan](https://selan.ai/blog/2026-10-17-cli-accounts): Stay signed in to every company you work for, and let a folder decide which one a Claude Code session spends as. Switching no longer means signing in again. Markdown: - [Shared connections: Reactor agents that act as a service account · Selan](https://selan.ai/blog/2026-10-17-shared-connections): A workspace owner signs in once as the account its agents should act as, and each agent picks it on its Connectors tab. Agents stop depending on one person's sign-in. Markdown: - [Agent groups: sort the Reactor canvas by the job each agent does · Selan](https://selan.ai/blog/2026-10-18-reactor-agent-groups): Put a workspace's agents in coloured boxes, one per job, by dragging their cards or by asking Claude through Selan MCP, so a canvas of twenty agents is no longer a wall of cards. Markdown: - [A ceiling on reasoning effort, per model maker · Selan](https://selan.ai/blog/2026-10-19-effort-ceiling): Claude Opus 5.5 at max effort can cost nine times what it costs at medium. An owner can now cap the effort for each model maker, and Logs show the effort each request ran at, where it named one. Markdown: - [Reactor agents now ask you when only a person can unblock them · Selan](https://selan.ai/blog/2026-10-24-agent-questions): A remote agent stuck on a revoked key or a missing access now asks you one short question. Answer in Reactor or in Slack, and its next run carries on from where it stopped. Markdown: - [Run OpenCode on your company's models, connectors and limits · Selan](https://selan.ai/blog/2026-10-27-opencode): selan opencode starts OpenCode on every model your company connected, with its connectors and skills and the same limits as Claude Code and Codex. Nothing goes into OpenCode's config. Markdown: - [Run Hermes Agent and Paperclip on your company's accounts · Selan](https://selan.ai/blog/2026-10-29-selan-apps): selan app installs and starts Hermes Agent or Paperclip on your company's models, with its limits and secret scanning. The catalog is public bash, and you read every line before it runs. Markdown: - [Remote agents you install with one line in Claude Code · Selan](https://selan.ai/blog/2026-10-30-install-remote-agents): Fifteen complete Reactor agents, one Markdown file each on GitHub. Paste one line into Claude Code, answer what is yours, and the agent is on your canvas. Markdown: - [Claude Code in JetBrains IDEs, on your company's accounts · Selan](https://selan.ai/blog/2026-10-30-jetbrains): One command sets up Claude Code in IntelliJ IDEA, GoLand, WebStorm, PyCharm and Android Studio, so a session started from the IDE runs through Selan like one from a terminal. Markdown: - [Run Kimi K3.1 Pro in Claude Code on your Kimi Code membership · Selan](https://selan.ai/blog/2026-10-30-kimi-code-subscription): Connect a Kimi Code membership on the Tokens page and Claude Code runs Kimi K3.1 Pro through Selan. An account without Kimi Code is now refused at sign-in, and Selan says which account to use. Markdown: - [Figma now works in Claude Desktop through Selan · Selan](https://selan.ai/blog/2026-10-31-figma-in-claude-desktop): Figma refuses Claude Desktop at sign-in. Selan now has its own Figma connector for Desktop: sign in once with a Figma token and Claude reads your files, frames, images and comments. Markdown: - [Virtual identities on your company's Google domain · Selan](https://selan.ai/blog/2026-10-31-virtual-identities): Create separate Google accounts for agents with Cloud Identity Free in your own Workspace. Set up licensing first; provider seats and permissions stay separate. Markdown: - [Selan Reactor is on Google Play: answer your agents from your phone · Selan](https://selan.ai/blog/2026-11-03-reactor-android): Selan Reactor for Android puts your workspaces on your phone. Answer an agent's question from its notification, stop a run, chat with the Expert or call Selan. Markdown: - [The Selan Chrome extension is now open to everyone · Selan](https://selan.ai/blog/2026-11-07-browser-extension-open-to-everyone): Add it to Chrome, point any MCP client at it, and your agent works in tabs where you are already signed in. You need no Selan account, and your agent need not be Claude. Markdown: ## Glossary The terms a team meets buying Claude Code, one definition each: `/glossary`. - [Claude Code glossary — seats, limits, keys, cache · Selan](https://selan.ai/glossary): The terms a team meets buying Claude Code: seat, gateway, provider key, usage and rate limits, prompt cache, MCP server, skill, plugin and subagent. Markdown: ## Top lists Rankings refreshed daily from the GitHub API, and a page per entry: `/top/`, `/top//this-week`, `/top//this-month`, `/top//-`. - [Top Claude Code skills: ranked from GitHub daily · Selan](https://selan.ai/top/claude-code-skills): Claude Code skills ranked daily by GitHub stars, and by this week's and this month's forks, commits and releases: what each does and how to install it. Markdown: - [addyosmani/agent-skills: install and activity · Selan](https://selan.ai/top/claude-code-skills/addyosmani-agent-skills): Addy Osmani's engineering skills for each stage of building software, from spec and plan to test, review and ship, behind nine slash commands. Markdown: - [ECC for Claude Code: install and activity · Selan](https://selan.ai/top/claude-code-skills/affaan-m-ecc): A plugin of 311 skills, 71 agents, hooks and rules that make an agent plan before it builds, test its changes, review its own work and keep what it learns. Markdown: - [anthropics/skills: install and activity · Selan](https://selan.ai/top/claude-code-skills/anthropics-skills): Anthropic's own skills for Claude, from the Word, PDF, PowerPoint and Excel skills behind its file creation to art, design, web app testing and MCP server building. Markdown: - [ponytail for Claude Code: install and activity · Selan](https://selan.ai/top/claude-code-skills/dietrichgebert-ponytail): One skill that makes the agent write only the code a task needs, without cutting validation, security or tests, plus commands to review a change and audit a repository. Markdown: - [Understand-Anything: install and activity · Selan](https://selan.ai/top/claude-code-skills/egonex-ai-understand-anything): A Claude Code plugin that builds a knowledge graph of a codebase, wiki or docs with a multi-agent pipeline and opens a dashboard to explore and search it. Markdown: - [gstack for Claude Code: install and activity · Selan](https://selan.ai/top/claude-code-skills/garrytan-gstack): Garry Tan's skills that make Claude Code act as an engineering team: product and architecture review, code review, QA in a real browser and a security audit, each a command. Markdown: - [caveman for Claude Code: install and activity · Selan](https://selan.ai/top/claude-code-skills/juliusbrussee-caveman): Makes an agent reply in clipped sentences and read less while code stays exact, and its README reports 33% fewer input tokens across whole Claude Code sessions. Markdown: - [taste-skill for Claude Code: install and activity · Selan](https://selan.ai/top/claude-code-skills/leonxlnx-taste-skill): Skills that give a coding agent an art direction, real photographs and a way to check its work in a browser, so the sites it designs stop looking average. Markdown: - [mattpocock/skills: install and activity · Selan](https://selan.ai/top/claude-code-skills/mattpocock-skills): Matt Pocock's everyday engineering skills, kept small and composable so you adapt them instead of handing a framework the whole process. Markdown: - [How the top Claude Code skills are ranked · Selan](https://selan.ai/top/claude-code-skills/method): How the top Claude Code skills list is made: where the numbers come from, how each tab ranks, and what is left out. Markdown: - [ui-ux-pro-max-skill: install and activity · Selan](https://selan.ai/top/claude-code-skills/nextlevelbuilder-ui-ux-pro-max-skill): A design skill that generates a design system for a project, choosing from its own catalogue of 79 UI styles, 192 colour palettes and 74 font pairings. Markdown: - [superpowers for Claude Code: install and activity · Selan](https://selan.ai/top/claude-code-skills/obra-superpowers): Composable skills that make a coding agent draw out a spec, write a plan, then build it test-first through subagents that review each task. Markdown: - [Agent-Reach for Claude Code: install and activity · Selan](https://selan.ai/top/claude-code-skills/panniantong-agent-reach): Sets up the command-line tools and a skill that let an agent read and search the web, X, Reddit, YouTube, Bilibili, XiaoHongShu, GitHub and RSS feeds. Markdown: - [impeccable for Claude Code: install and activity · Selan](https://selan.ai/top/claude-code-skills/pbakaus-impeccable): A design skill that gives a coding agent 24 commands and 62 detector rules to steer its frontend work away from generic AI-looking design. Markdown: - [claude-mem for Claude Code: install and activity · Selan](https://selan.ai/top/claude-code-skills/thedotmack-claude-mem): Captures what Claude Code does in each session as compressed summaries and makes them available to later sessions, so context carries over. Markdown: - [Trending Claude Code skills: this month on GitHub · Selan](https://selan.ai/top/claude-code-skills/this-month): The Claude Code skills gaining the most forks, commits and contributors on GitHub in the last 30 days. Updated daily. Markdown: - [Trending Claude Code skills: this week on GitHub · Selan](https://selan.ai/top/claude-code-skills/this-week): The Claude Code skills gaining the most forks, commits and contributors on GitHub in the last 7 days. Updated daily. Markdown: - [archify for Claude Code: install and activity · Selan](https://selan.ai/top/claude-code-skills/tt-a1i-archify): A skill that turns a description, a plan or a repository into an interactive HTML diagram you can explore, change and share. Markdown: