# Cite Files

> Check whether AI assistants can reach, read and cite a public website. No account for 4 of 5 tools.

- name: com.citefiles/citefiles
- version: 1.0.0
- connection class: R0 (autonomous)
- trust: 82/100 flags: no-repository
- quality: 88/100 (good)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/com.citefiles%2Fcitefiles
- tools: https://api.protogrid.dev/v1/servers/com.citefiles%2Fcitefiles/tools

**An agent can connect right now, no human step.** Remote endpoint, no authentication, reachable on the last probe.

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "citefiles": {
      "type": "http",
      "url": "https://citefiles.com/mcp"
    }
  }
}
```


## Trust

- hygiene: 80
- liveness: 99
- freshness: 100
- provenance: 55
- drivers: +dns-namespace +website +reachable +uptime-97% +updated-16d-ago -no-repository
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 88/100 (good)

### protocol: 100 (weight 25)

- pass `protocol.modern`: supports 2026-07-28 (server/discover)
- pass `protocol.stateless`: answers without a session
- pass `protocol.transport`: streamable HTTP
- pass `protocol.list_ttl`: tool list is cacheable (ttlMs 3600000)

### authorization: n/a (weight 15)

- n/a `auth.prm`, `auth.as_metadata`, `auth.cimd`: no remote uses OAuth
- n/a `auth.secret_in_url`: no templated URL

### tool hygiene: 71 (weight 30)

- pass `tools.descriptions`: every tool has a description
- pass `tools.description_length`: descriptions are concise
- pass `tools.schemas`: every tool has an object input schema
- fail `tools.annotations`: no tool declares readOnlyHint or destructiveHint, so clients cannot tell safe calls from risky ones
- fail `tools.directory_hints`: no tool declares all four of readOnlyHint, destructiveHint, idempotentHint and openWorldHint; missing: citefiles_fix_prompt (readOnlyHint, destructiveHint, idempotentHint, openWorldHint); citefiles_check_files (readOnlyHint, destructiveHint, idempotentHint, openWorldHint); citefiles_crawler_access (readOnlyHint, destructiveHint, idempotentHint, openWorldHint); …
- pass `tools.token_cost`: about 785 tokens to load every tool
- pass `tools.api_dump`: tools are not a one-to-one API dump

### stability: 100 (weight 15)

- pass `stability.changes`: 1 tool changes in 30 days
- pass `stability.rug_pull`: no tool changed its meaning under the same name

### dependencies: n/a (weight 15)

- n/a `deps.known_vulns`, `deps.mcp_sdk_version`, `deps.resolvable`: no npm or PyPI package

- tools: 5, about 785 tokens to load them all
- tool set hash: 862a7a75ac088e6a8e21bbf6a4cce12d
- tools last changed: 2026-10-04T16:43:50.350Z
- badge: [![protogrid quality](https://protogrid.dev/badge/com.citefiles/citefiles.svg)](https://protogrid.dev/servers/com.citefiles/citefiles)
- Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed.

### Tool changes

- 2026-10-04 `citefiles_crawler_access` changed (description)
  - before: Test how a website answers the real AI crawlers — GPTBot, ClaudeBot, PerplexityBot and Google-Extended — by requesting its homepage as a browser and again as each crawler, then comparing. Catches CDN and WAF rules that block assistants without the owner knowing. Use when a site looks correct but is not being cited.
  - after: Test how a website answers requests that carry the user-agent strings of four AI crawlers — GPTBot, OAI-SearchBot, ClaudeBot and PerplexityBot — by requesting its homepage as a browser and again as each, then comparing. The requests come from our server, not the vendors' networks, so a rule that checks the sender's address is not tested. Catches CDN and WAF rules that block assistants without the owner knowing. Use when a site looks correct but is not being cited.

Full history: https://api.protogrid.dev/v1/servers/com.citefiles%2Fcitefiles/changes

## Remotes

- https://citefiles.com/mcp (streamable-http, auth none, reachable true, uptime30d 0.96825397)

## Packages

_none_

## Tools (5)

- `citefiles_agent_commerce`: Requires an API token (Authorization: Bearer cf_…). Read the stored Agent Commerce Readiness audit for a scan on your citefiles account: whether an AI agent can authenticate without a human's password, discover an API, read prices as text, and be offered a machine-payable challenge — as a tier from T0 to T5. Create a token at https://citefiles.com/account. Only returns audits for scans your account owns. This tool never runs a new audit — run it from the scan's report page first, then call this tool to read the result.
- `citefiles_check_files`: Check which AI-facing discovery files a website actually serves: llms.txt, llms-full.txt, ai.txt, robots.txt, sitemap.xml, AGENTS.md, .well-known/mcp.json and others. Distinguishes a real file from a single-page app returning its HTML shell for every unknown path, which is the usual reason a file appears to exist but does not. Use this to verify a file you just published is being served correctly.
- `citefiles_crawler_access`: Test how a website answers requests that carry the user-agent strings of four AI crawlers — GPTBot, OAI-SearchBot, ClaudeBot and PerplexityBot — by requesting its homepage as a browser and again as each, then comparing. The requests come from our server, not the vendors' networks, so a rule that checks the sender's address is not tested. Catches CDN and WAF rules that block assistants without the owner knowing. Use when a site looks correct but is not being cited.
- `citefiles_fix_prompt`: Get a task list for making a website readable and citable by AI assistants, derived from a real scan of it. Returns the findings with the evidence behind each, and the rules to follow while fixing them — most importantly that you must not invent facts about the business, and must leave a visible TODO instead. Use this to plan the work after citefiles_scan.
- `citefiles_scan`: Check whether AI assistants can reach, read and cite a public website. Returns a citation-readiness score out of 100, the category breakdown, and the specific problems found — crawler blocks, missing llms.txt/ai.txt/sitemap, thin or JavaScript-only content, missing structured data. Use this before changing a site for AI visibility, and again afterwards to check the change worked. Scans of the same site within six hours are reused.