# Cited (pypi · cited-mcp)

Measure and improve how often AI assistants cite your business: GEO audits, fixes, and docs.

- Trust score: 65/100 (medium)
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- remote · `mcp.youcited.com`: 38/100, [markdown](https://verifymcp.io/servers/com-youcited-cited/mcp.md), [page](https://verifymcp.io/servers/com-youcited-cited/mcp)
- pypi · `cited-mcp`: 65/100 (this document), [markdown](https://verifymcp.io/servers/com-youcited-cited/cited-mcp.md), [page](https://verifymcp.io/servers/com-youcited-cited/cited-mcp)

## Channel facts

- Registry: `pypi`
- Package: `cited-mcp`
- Version: `0.8.1`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-05.

- **Supply Chain Security**: 99/100
  - No malware found by supply-chain analysis.
  - No known CVEs affecting this package version or its production dependencies.
  - Runs hatchling.build at install time, a recognised build step with no custom scripting around it.
  - 5 of 47 dependencies flagged as unhealthy.
- **Provenance & Transparency**: 32/100
  - Source repository is publicly reachable at the declared URL.
  - Build provenance not yet verified.
  - License check failed: the license (LicenseRef-Proprietary) isn't a recognized OSI-approved license.
  - Actively maintained (last published 6 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 65/100
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 2733 tokens (~130/item across 21 items; 21 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 0/100
  - Stability not yet verified: not enough scan history yet (needs a 30-day window).
- **Tool Coverage**: 89/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 68% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 21 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 22 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a current MCP spec version (2026-07-28).

**Unverified: 1 category.** A category scored 0 because we could not verify it: a data source with nothing on this package, evidence we could not reach, or a check we could not run. We only credit what we can confirm.

## Install

### How do I install the Cited MCP server?

Cited runs locally as a PyPI package, launched with uvx cited-mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add com-youcited-cited -- uvx cited-mcp
```

### Cursor

```json
{
  "mcpServers": {
    "com-youcited-cited": {
      "command": "uvx",
      "args": [
        "cited-mcp"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "com-youcited-cited": {
      "command": "uvx",
      "args": [
        "cited-mcp"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add com-youcited-cited -- uvx cited-mcp
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "com-youcited-cited": {
      "type": "local",
      "command": [
        "uvx",
        "cited-mcp"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add com-youcited-cited --command uvx --arg cited-mcp
```

### Hermes

```yaml
mcp_servers:
  com-youcited-cited:
    command: "uvx"
    args: ["cited-mcp"]
```

### Netclaw

```json
{
  "McpServers": {
    "com-youcited-cited": {
      "Transport": "stdio",
      "Command": "uvx",
      "Arguments": [
        "cited-mcp"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add com-youcited-cited -t stdio -c uvx -a cited-mcp
```

### Other

```json
{
  "mcpServers": {
    "com-youcited-cited": {
      "command": "uvx",
      "args": [
        "cited-mcp"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-05 (score 65, +15)

- [security regression] Provenance: fail → unverified
- [security improvement] Malware scan: unverified → pass

### 2026-11-03 (score 50)

First indexed and scored.

## MCP tools (21)

### `free_audit` (~150 tokens)

Free AI Visibility Audit

Free check of how AI answers 5 buyer questions about a website: whether the site is cited, which sites are cited instead, a 0-100 score, an executive summary and whether AI bots can reach the site. No account needed. The first call starts it and returns report_token; call again with report_token until status is 'completed' (2-4 minutes; each call waits up to 20 seconds, so no need to pause between calls). The same domain within 24 hours returns the same audit.

Input parameters:

- `report_token` (string): From an earlier free_audit call, to check on it.
- `url` (string): Website to audit, e.g. example.com.

### `agent_signup` (~187 tokens)

Create Account For Your Human

Create a Cited account for your human without a login. Returns an API key (shown once), a claim link and a 6-digit code for the human. Unclaimed accounts can add 1 business and run 1 audit (paid with a one-audit checkout link) and are deleted after 7 days. Pass owner_email to email them the claim link. Pass free_audit_token (the report_token of a finished free_audit) to bring that business and report into the new account. Not needed when already signed in.

Input parameters:

- `agent_name` (string): Your name, shown to the human (e.g. 'Claude').
- `free_audit_token` (string): Optional: the report_token of a finished free_audit, to keep that business and report in the new account.
- `owner_email` (string): Optional: the human's email for the claim link.

### `account_status` (~49 tokens)

Account Status

Who is signed in, the credit balance, claim status, the businesses on the account, the price list and the next step. Works without signing in (prices and how to start). Call this first.

### `buy_credits` (~153 tokens)

Buy Credits

Get a Stripe checkout link for a credit pack (pack_id) or for one audit (one_audit=true; works before the account is claimed), or redeem a voucher_code. Without arguments it lists the packs. A link only starts a purchase: the human pays in the browser, and nothing is bought until account_status shows the new balance. 1 credit = $1.

Input parameters:

- `one_audit` (boolean): Checkout for a single audit instead of a pack.
- `pack_id` (string): A pack id from the price list (account_status).
- `save_card` (boolean): Save the card for later packs (default true).
- `voucher_code` (string): Redeem a voucher for free credits.

### `setup_business` (~343 tokens)

Set Up Business

Add a business from its website, or update one (pass business_id). Only website is needed to add one: the name comes from the domain and industry defaults to 'other' unless you pass better ones. Also adds, changes or removes buyer personas and offerings (what the business sells). Unclaimed accounts can add one business.

Input parameters:

- `address` (string): Update only.
- `business_id` (string): Update this business instead of adding one.
- `city` (string): Update only.
- `country` (string): Update only.
- `description` (string): What the business does (10+ words, 50+ characters).
- `discovery_channels` (array)
- `goals` (array)
- `headquarters_country` (string): Two-letter ISO code, e.g. US.
- `industry` (string)
- `name` (string)
- `offering_type` (string)
- `offerings` (array): Products or services to add ({name, description, category}), change ({id, ...}) or remove ({id, remove: true}).
- `one_line_description` (string): Tagline, max 140 characters.
- `personas` (array): Buyer personas to add ({name, description}), change ({id, ...}) or remove ({id, remove: true}).
- `primary_customer` (string)
- `primary_market` (string)
- `problem_solved` (string)
- `region` (string): Update only.
- `service_areas` (string): Cities or regions, for local or regional businesses.
- `success_definition` (string)
- `website` (string): e.g. example.com

### `get_business` (~51 tokens)

Get Business

Without business_id: list the account's businesses. With it: the profile, buyer personas, offerings, the question set (and its re-check schedule) and recent audits.

Input parameters:

- `business_id` (string)

### `get_agent_brief` (~56 tokens)

Agent Brief

One-page summary of what to focus on for a business: top open recommendations, quick wins, failing site checks and recent citation movement. Use it for 'what should I do next?'.

Input parameters:

- `business_id` (string, required)

### `draft_questions` (~200 tokens)

Draft Audit Questions

The business's question set: the buyer questions each audit asks AI. Returns the set when there is one. With none (or regenerate=true, or add=true to draft more) it starts drafting and returns job_id; call again with job_id to save and return the drafted set. Pass questions to replace the set with your own list. Review the questions with the user before run_audit. Up to 25 questions.

Input parameters:

- `add` (boolean): Draft more questions and add them to the set.
- `business_id` (string): Optional when the account has one business.
- `count` (integer): How many to draft (default 5).
- `focus` (string): Optional steer, e.g. 'pricing questions'.
- `job_id` (string): From an earlier call, to collect drafted questions.
- `questions` (array): Replace the set with these questions.
- `regenerate` (boolean): Draft a new set even though one exists.

### `quote_audit` (~68 tokens)

Quote Audit

What an audit of the business's question set costs in credits, the balance, and the models that can answer. Free; nothing is charged.

Input parameters:

- `business_id` (string): Optional when the account has one business.
- `models` (array): Optional models to compare, from available_models.

### `run_audit` (~202 tokens)

Run Audit

Run a paid audit of the business's question set: every question goes to each engine's flagship and fast model (or the models you pass), three times each, and recommendations follow automatically. Spends credits: without confirm=true it only returns the cost, so ask the user, then call again with confirm=true. Credits come back if the run fails. Returns audit_id; poll get_audit.

Input parameters:

- `business_id` (string): Optional when the account has one business.
- `confirm` (boolean): True only after the user approved the cost.
- `country` (string): Two-letter country for as-seen (brightdata/) checks.
- `deep_recommendations` (boolean): Look harder at why other sites were cited.
- `engines` (array): Optional subset of engines (default: all).
- `models` (array): Optional models to compare, as [route/]provider:model from quote_audit's available_models. Each extra model costs more.

### `get_audit` (~129 tokens)

Get Audit

With audit_id: its status while it runs, then the report: a 0-100 score per model (Cited, Visibility, Share of Voice), the overall score, the executive summary, the site check, each question's answers with the sites cited instead, the share link and whether recommendations are ready. Without audit_id: the account's audits (filter by business_id).

Input parameters:

- `audit_id` (string)
- `business_id` (string)
- `include_answers` (boolean): Full AI answers instead of excerpts.
- `limit` (integer): List length (default 20).

### `compare_audits` (~68 tokens)

Compare Audits

Wins and drops between an audit and an earlier one of the same business: citations gained or lost per question and model, and score changes. Without baseline_audit_id it compares with the previous audit.

Input parameters:

- `audit_id` (string, required)
- `baseline_audit_id` (string)

### `schedule_recheck` (~137 tokens)

Schedule Re-check

Re-run the business's question set on a schedule (every 7-90 days, default 30) and alert on citations gained or lost and score moves of 10+ points: in-app, by email, and to an optional webhook so an agent can react. Each re-check costs credits when it runs. off=true stops it.

Input parameters:

- `business_id` (string): Optional when the account has one business.
- `email_alerts` (boolean)
- `every_days` (integer)
- `off` (boolean): Stop the schedule.
- `webhook_url` (string): HTTPS URL for alerts; empty string removes it.

### `get_recommendations` (~147 tokens)

Get Recommendations

What to change on the site to get cited, from an audit: one item per buyer question the site loses and per site change, each with item_key, status (open, done, dismissed), the evidence behind it and its fix if one was made. Open items only unless include_closed. Also the price of fixes (the first ones are included with each audit).

Input parameters:

- `audit_id` (string): The audit (default: the newest finished one).
- `business_id` (string): Use this business's newest finished audit.
- `include_closed` (boolean): Also list done and dismissed items.
- `recommendations_id` (string): From get_recommendations; overrides audit_id.

### `update_recommendation` (~104 tokens)

Update Recommendation

Mark a recommendation done or dismissed, or open it again. The status is kept for the business, so later audits show the item the same way.

Input parameters:

- `audit_id` (string): The audit (default: the newest finished one).
- `business_id` (string): Use this business's newest finished audit.
- `item_key` (string, required)
- `recommendations_id` (string): From get_recommendations; overrides audit_id.
- `status` (string, required)

### `generate_fixes` (~171 tokens)

Generate Fixes

Write the fix for recommendations: the files (schema markup, FAQ, page copy, llms.txt) and steps to put them on the site. Pass item_keys (up to 10) or all=true for every open item without a fix. Each audit includes the first fixes; beyond those each costs credits, and the call returns the price for the user to confirm (confirm=true) before spending.

Input parameters:

- `all` (boolean): Every open item.
- `audit_id` (string): The audit (default: the newest finished one).
- `business_id` (string): Use this business's newest finished audit.
- `confirm` (boolean): The user agreed to the credits quoted.
- `item_keys` (array)
- `recommendations_id` (string): From get_recommendations; overrides audit_id.

### `get_fixes` (~133 tokens)

Get Fixes

With fix_id: the fix's files (with their content), where each goes, the steps and the last verification. Without: list fixes with their state (generating, ready, done, verified, working), optionally for one audit, recommendations run or business.

Input parameters:

- `audit_id` (string): The audit (default: the newest finished one).
- `business_id` (string): Use this business's newest finished audit.
- `fix_id` (string)
- `limit` (integer)
- `recommendations_id` (string): From get_recommendations; overrides audit_id.
- `state` (string)

### `update_fix` (~78 tokens)

Update Fix

action 'done': the user put the fix on their site. 'undo': back to ready. 'verify': re-check the live site for the fix now (free); it moves to verified when every check passes, and the report says what is missing when one fails.

Input parameters:

- `action` (string, required)
- `fix_id` (string, required)

### `search_docs` (~78 tokens)

Search Cited Docs

Search the Cited help docs (how audits, scores, credits, fixes and the tools work) with query, or read one doc in full with path ('<category>/<slug>' from a result). No account needed.

Input parameters:

- `limit` (integer)
- `path` (string): A result's path.
- `query` (string)

### `login` (~93 tokens)

Log In

Sign in to Cited with a browser. The first call opens the sign-in page and returns login_url to show the user; once they have signed in, call it again to finish. force=true signs in again (to switch accounts). Not needed for free_audit, account_status or search_docs, and agent_signup makes an account without a login.

Input parameters:

- `force` (boolean): Sign in again even when already signed in.

### `logout` (~32 tokens)

Log Out

Sign out on this machine: forget the stored session, or the agent key saved by agent_signup when there is no session.

## Diagnostics

Captured diagnostic sections: Provenance, Install scripts, Dependencies. The full working is on the page: https://verifymcp.io/servers/com-youcited-cited/cited-mcp#diagnostics

## Score history

- 2026-11-05: 65
- 2026-11-04: 50
- 2026-11-03: 50

## Common questions

### What is the Cited MCP server?

Cited is an MCP server listed in the public MCP registry as com.youcited/cited. Measure and improve how often AI assistants cite your business: GEO audits, fixes, and docs. This page covers its PyPI package (cited-mcp).

### Is the Cited MCP server safe to use?

Cited scores 65 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 5 November 2026. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the Cited MCP server expose?

Cited exposes 21 tools: free_audit, agent_signup, account_status, buy_credits, setup_business, and 16 more. Their descriptions and schemas cost roughly 2,629 tokens of context every time the server is loaded.

### Is the Cited MCP server still maintained?

Cited is still listed as active in the MCP registry. We last reached this channel on 5 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What licence is the Cited MCP server under?

Cited declares the LicenseRef-Proprietary licence, which is not on the OSI-approved list. Read the terms before using it at work, and note this covers the source only, not the cost of any service it calls.

## Links

- PyPI project: https://pypi.org/project/cited-mcp/
- Socket report: https://socket.dev/pypi/package/cited-mcp
- Repository: https://github.com/YouCited/cited-cli
- Website: https://youcited.com/
- Changelog RSS feed: https://verifymcp.io/servers/com-youcited-cited/cited-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/com-youcited-cited/cited-mcp.json
- HTML version of this page: https://verifymcp.io/servers/com-youcited-cited/cited-mcp