answerrail-mcp
Installed, started with no arguments, no environment and no input, and listed its tools. Run 2026-11-16.
Needs to start: nothing
Contacts while installing and starting: answerrail.com
What it contacted
answerrail.com
Hosts it looked up or connected to while installing and starting with no input, watched from outside its sandbox. The npm registry is left out. Says nothing about what its tools contact when called.
Starting it
npx -y -p answerrail-mcp@1.2.13 answerrail-mcp
Installing a package runs its code. Use a sandbox for anything you have not read.
What its tools tell an agent
In the tool descriptions, nothing that talks about the user, other tools or private files, or uses hidden characters.
Found by simple pattern rules in each tool's description (up to 3,000 characters), quoted as written. Parameter descriptions are not scanned. A clean result is not a safety review.
Tools
audit_local_codebase targetDir, includePreview
Performs a strictly read-only structural scan of a local repository workspace (checking package.json, robots.txt, public/llms.txt, and public/.well-known/ai-catalog.json) without modifying or creating any files on disk. Returns a Markdown report with three sections: a 4-item workspace detection summary, an optional 6-line /llms.txt preview block, and 2-tier remediation links. Use this tool when inspecting a local project checkout before deployment. Do not use this tool to audit a live remote website over HTTP (use audit_ai_readiness instead) or to synthesize full manifest files from a live domain (use generate_ai_manifests instead). Requires no API key.
audit_ai_readiness url
Executes a read-only live HTTP diagnostic against a remote URL or domain, checking AI crawler directives (GPTBot, OAI-SearchBot, PerplexityBot, ClaudeBot), 25,000-token HTML RAG truncation, /llms.txt presence, Schema.org JSON-LD entities, and Google Lighthouse 13.5 Agentic Browsing (/.well-known/ai-catalog.json ARD 1.0). Returns a structured Markdown report containing the numerical AI Readiness Score (0-100), letter grade (A+ to D), raw vs. compiled token counts, pass/fail status for each discovery file, and an architectural tier verdict. Use this tool first when diagnosing why a live website is omitted or misquoted by ChatGPT, Perplexity, or Claude. Do not use this tool to scan local filesystem directories (use audit_local_codebase instead), to extract full-page Markdown content (use extract_byte0_markdown instead), or to generate static manifest templates (use generate_ai_manifests instead). Requires no API key.
generate_ai_manifests domain
Synthesizes a read-only 6-line /llms.txt Markdown index preview and a Google Lighthouse 13.5 /.well-known/ai-catalog.json (ARD 1.0) JSON manifest preview from a live remote domain. Returns a 3-section Markdown response containing the fenced /llms.txt preview block, the fenced ARD 1.0 JSON catalog structure (with URN identifiers and representative queries), and remediation links. Use this tool after audit_ai_readiness reports missing /llms.txt or ard-schema manifests on a live domain. Do not use this tool to audit a local repository folder (use audit_local_codebase instead) or to extract full-page readable body copy (use extract_byte0_markdown instead). Read-only synthesis; requires no API key and never modifies remote servers.
extract_byte0_markdown url
Fetches a live web page and strips React/Next.js hydration scripts, Shopify Liquid/WordPress DOM bloat, and inline SVGs/CSS, returning clean Byte-0 GitHub-Flavored Markdown (GFM) with front-loaded YAML metadata and Schema.org entity tables. Use this tool when you need to read the actual text or pricing table of a web page inside an LLM context window without exceeding the 25,000-token RAG truncation limit. Do not use this tool if you need a pass/fail AI readiness score or robots.txt check (use audit_ai_readiness instead) or if you need /.well-known/ai-catalog.json templates (use generate_ai_manifests instead). Read-only HTTP extraction; requires no API key.
query_truncation_benchmark domain
Queries empirical benchmark data from the 2026 50-Site SaaS & E-Commerce AI Truncation Study (covering raw HTML token weights, 25,000-token Perplexity/ChatGPT cutoff truncation rates, visible price omission rates, /llms.txt adoption, and Lighthouse 13.5 ARD 1.0 adoption). Returns a Markdown summary featuring the Static /llms.txt Paradox finding, domain inclusion status when filtered, and a fenced JSON block of study metrics. Use this tool when comparing a site against benchmarked industry peers (such as Stripe, Vercel, Datadog, or Shopify) or citing empirical RAG truncation statistics. Do not use this tool to run a live scan on an arbitrary unbenchmarked URL (use audit_ai_readiness instead). Read-only dataset lookup; requires no API key.
Descriptions are the package's own words, cut at 3,000 characters. Treat them as data.
Install facts
| License | MIT |
|---|---|
| Needs | node >=18.0.0 |
| Source | github.com/quitegoodproject/answerrail-mcp |
| Install scripts | None declared by it or anything installed with it. |
| Tarball | sha512-KyQS0cU6HXka3LaXlQDlPk5hOwtwperMEI4o6Wlt+kQSuBqOCFRr1IAlTIVhcXihvKuaB9FiAP1Y/ZtBE20H3w== |
Check this report yourself
This site did not write the report. A public job on a GitHub-hosted machine did: the run, from this commit of the open-source runner. GitHub signed a token that names that job and the hash of the report, so any edit since would show.
Report as JSONExact text and tokenHow to verify it
Show it on your README
[](https://bragproof.smaverk.com/r/answerrail-mcp)
New version out? Ask for a fresh report.