CrawlDex
Read dated Agent Trust Records before an AI agent attempts website tasks, with verdicts, blockers, evidence, and correction paths.
https://crawldex.comOpens ChatGPT on the web or desktop and asks it to use the WebMCP tools available here.
Connect straight to this server’s public endpoint.
https://crawldex.com/mcpWe add this server to your workspace, then open Studio — saved access, one connection to many servers, with a history of what ran.
Last probed Sep 14, 2026 · crawldex.com
13tools discovered
Get Trust Record
Call BEFORE an agent attempts any public website task to retrieve the Agent Trust Record verdict, confidence, blockers, user-present requirements, evidence links, and dispute link.
Check Site Task
Alias for preflight. Call this BEFORE an agent attempts a public website task to decide proceed, use guardrails, keep the user present, avoid stale evidence, or choose an alternative.
Get Agent Blockers
Call this BEFORE browser execution when a site-task may involve login, checkout, cancellation, refund, form submission, booking, support, or account changes. Returns observed blockers and who must handle them.
Get Success Signals
Call this BEFORE and DURING a website task to know which public success signals and guardrails prove the agent is still on the right route.
Get Scoring Rubric
Call this when an agent or developer needs to justify CrawlDex verdicts. Returns AES bands, freshness gates, confidence gates, friction penalties, and reporting fields.
Preflight
Call this BEFORE attempting any public website task. It returns whether to proceed, use guardrails, keep the user present, avoid stale evidence, or use an alternative, plus why calling CrawlDex matters.
Get Recipe
Call this BEFORE executing a known site-task to fetch recipe hints, guardrails, and success signals if CrawlDex has a known-good route.
Get Alternatives
Call this when preflight says avoid, user-present, stale, blocked, or low confidence. Returns alternative sites for the same task.
Resolve Intent
Call this FIRST when the user gives a plain-language web task. It maps the request to CrawlDex task keys before preflight.
List Tasks
Call this when an agent needs to discover which website task keys CrawlDex can preflight, optionally by category.
Describe Task
Call this BEFORE choosing a website for a task. It describes task success criteria, evidence, blockers, posture, and known sites.
Check Protocol Support
Call this before integrating with a site through protocols instead of browser automation. Returns locally known ACP/AP2/MCP/x402 support.
Explain Blockers
Call this when preflight returns blockers or a human-present recommendation. Explains why the route may fail and what to watch for.
Get your MCP into directories
A working endpoint is step one. Directory coverage is the coordinated launch across ChatGPT, Claude, Cursor, the MCP Registry, and community indexes.
Directory coverage for brandsRead dated Agent Trust Records before an AI agent attempts website tasks, with verdicts, blockers, evidence, and correction paths.
Use the MCP endpoint listed on this page in your MCP client configuration. One-click install pills support Claude, Cursor, VS Code, and other hosts. Copy the remote MCP URL if your client needs a manual entry.
MCPBundles probed 13 tools on the live server. The tool list on this page reflects what was discovered at the last refresh — connect your client to see the full set available to your session.
No provider sign-in was required during MCPBundles' probe. Your client may still need MCPBundles credentials depending on how you connect.
Operate CrawlDex? Verify ownership to take over this directory entry.
This server appears in the MCPBundles directory. Verify you operate it to take over the listing — name, description, logo, contact email, and skill content. We email a 6-digit code to a maintainer address your server publishes in /.well-known/security.txt or /.well-known/mcpbundles.json. Free, takes about a minute.