One call tells your AI agent whether a website will block it (bot challenges, CAPTCHAs, robots.txt rules for GPTBot, ClaudeBot and others) and where the site's agent door is: llms.txt, MCP, or OpenAPI.
Each result includes understood, job, and a one-line outcome, so your agent can confirm and report back in plain words.
check_agent_accessBefore an agent browses or fetches a site, check if it will get in. Fetches the URL as an AI agent and as a browser, detects bot challenges (Cloudflare, CAPTCHA, DataDome, PerimeterX), reads robots.txt rules for GPTBot/ChatGPT-User/ClaudeBot/Claude-User/PerplexityBot and others, and looks for agent doors (llms.txt, MCP, OpenAPI).
find_agent_alternativeFind a machine-friendly way into a site instead of scraping its HTML: llms.txt, /.well-known/mcp.json, agents.json, openapi.json, ai-plugin.json, and MCP URLs listed in llms.txt. Read-only.
check_manyRun check_agent_access on up to 10 URLs at once (counts as 1 call). Use it to pick which sources an agent can actually use.
usage_and_upgradeUsage today + upgrade link (not metered).
The free tier resets every day. It is not a trial. Upgrade when your agents need volume.
No signup. 50 calls a day per IP over MCP or REST.
A stable identity for your agent with the same 50/day. One click, no email.
5,000 calls a day. After you pay, you get your key immediately. Cancel anytime.
It tells an AI agent, in one call, whether a website will let it in: it fetches the page as an agent and as a browser, detects bot challenges and CAPTCHAs, reads robots.txt rules for AI user agents, and lists the site's agent doors (llms.txt, MCP, OpenAPI).
Yes. 50 checks per day are free with no signup, every day. Pro is $9/month for 5,000 checks per day.
Add the remote MCP server https://accesscheck.botrecipe.com/mcp. In Claude Code run: claude mcp add --transport http agent-access-check https://accesscheck.botrecipe.com/mcp. Or call the REST API at /api/run/check_agent_access.
GPTBot, ChatGPT-User, OAI-SearchBot, ClaudeBot, Claude-User, PerplexityBot, Google-Extended, CCBot and the * default group.
No. It only detects them and points the agent to an allowed alternative such as llms.txt, an MCP server or an API. It is read-only and respects robots.txt.
No. There is no model inside, so it is fast and deterministic. Bring your own model.
Remote MCP over streamable HTTP. Claude Code:
claude mcp add --transport http agent-access-check https://accesscheck.botrecipe.com/mcp
MCP URL https://accesscheck.botrecipe.com/mcp
REST POST https://accesscheck.botrecipe.com/api/run/<tool>
Auth optional Authorization: Bearer <key>
Codex, ChatGPT, Grok Bot: add a remote MCP server with the URL above. Skill file: SKILL.md · catalog entry: tool.json