CitePatch

AI visibility for marketing teams, exposed to your agent.

CommunityOnlineNo sign-inGlobalFreeCan modify data

What it can do

  • Citepatch Apply Patch: Apply an approved patch via connector (opens a reviewable PR/draft). Org mcp_apply_enabled opt-in required. First call without confirmed returns confirmation_required + one-use
  • Citepatch Get Brand Truth: Returns approved Brand Truth facts for a project. Requires project_id. Facts are verified only when approved in CitePatch.
  • Citepatch Get Evidence For Gap: Returns the real evidence chain for a gap. Requires project_id and playbook_id, where playbook_id is the stable ID from actionable_gaps (internally the opportunity ID).

What data it sees

Do you need an account

No: the server works without sign-in

AI visibility for marketing teams, exposed to your agent. Read the evidence behind how AI assistants answer your buyers' questions (visibility summary, prompt answers, cited pages, competitors), the opportunities and playbooks CitePatch found, Brand Truth, and the reviewable patches — then submit or apply a patch with human approval. OAuth 2.1 with per-scope consent; nothing is published without a person saying yes. A read-only endpoint is available at https://mcp.citepatch.com/mcp/readonly. Requires a paid CitePatch plan. Docs: https://citepatch.com/mcp/

How to connect

How to connect CitePatch to Claude

The server address to paste into Claude settings:

https://citepatch--support-q2pb.run.tools

  1. Open Claude (claude.ai or the desktop app).
  2. Go to Settings → Connectors.
  3. Click Add custom connector.
  4. Paste the server address copied below into Remote MCP server URL and click Add.
  5. In a chat, click + → Connectors and switch the new connector on.

Custom connectors are available on Free, Pro, Max, Team and Enterprise plans (Free is limited to one). On Team and Enterprise an organization Owner adds the connector first under Organization settings → Connectors.

Authorization

After you click Add, a sign-in window for the service opens. Sign in with your own account and approve access. Claude never sees your password.

How to connect CitePatch to ChatGPT

The server address to paste into ChatGPT settings:

https://citepatch--support-q2pb.run.tools

  1. Open ChatGPT in a browser (chatgpt.com). A Plus, Pro, Business, Enterprise or Edu plan is required.
  2. Turn on developer mode once: Settings → Apps → Advanced settings → Developer mode.
  3. Open Settings → Connectors and click Create.
  4. Fill in the form: Name (anything), Description (one line about what the service does), MCP server URL (copy it below).
  5. Under Authentication choose OAuth if the service requires sign-in, otherwise None. Click Create.
  6. In a new chat open + → Apps/Connectors and enable the connector.

OpenAI has renamed this section before (Connectors → Apps/Plugins). If the label differs, search settings for "developer mode". On Business/Enterprise workspaces an admin must allow custom connectors first.

Authorization

On first use ChatGPT opens the service's sign-in window. Sign in and approve access.

How to connect CitePatch to Cursor

The server address to paste into Cursor settings:

https://citepatch--support-q2pb.run.tools

Fastest: click Open in Cursor below and confirm the prompt.

Manually:

  1. In Cursor open Settings → Cursor Settings → MCP and click Add new global MCP server.
  2. Paste the JSON copied below into ~/.cursor/mcp.json (per project: .cursor/mcp.json in the repo root).
  3. Save the file. The server appears in the MCP list; authorize it there if asked.

Authorization

If the service needs sign-in, an authorize button appears next to the server in the MCP list.

Server tool list (14)

Raw names from tools/list. Only developers need these.

citepatch_apply_patchApply an approved patch via connector (opens a reviewable PR/draft). Org mcp_apply_enabled opt-in required. First call without confirmed returns confirmation_required + one-use handle; second call with confirmed=true executes. Audit is fail-closed before the connector write.
citepatch_get_brand_truthReturns approved Brand Truth facts for a project. Requires project_id. Facts are verified only when approved in CitePatch.
citepatch_get_evidence_for_gapReturns the real evidence chain for a gap. Requires project_id and playbook_id, where playbook_id is the stable ID from actionable_gaps (internally the opportunity ID). The chain links prompt_id -> gap_identity -> evidence snapshots; playbook_state=missing means no playbook exists. Prefer after citepatch_list_prompts or citepatch_list_opportunities.
citepatch_get_experimentReturns experiment ledger entries for a project. Requires project_id; optional experiment_id filters to one experiment.
citepatch_get_opportunityReturns one opportunity by id for a project. Requires project_id and opportunity_id.
citepatch_get_patchReturns patch details for a project. Requires project_id; optional patch_id filters to one patch.
citepatch_get_patch_briefReturns the versioned authorship brief (brief_version) for an owned approved opportunity: allowed paths, citable brand_fact_ids, prohibitions, validation battery, and submit schema. Requires project_id and opportunity_id.
citepatch_get_project_briefingReturns project status briefing (name, domain, onboarding, panel, last measurement). Requires project_id. Do not use for metrics time series — use citepatch_get_visibility_summary.
citepatch_get_prompt_answersReturns observed AI answers for a prompt_id. External model text is fenced as untrusted data. Requires project_id and prompt_id. Prefer citepatch_list_prompts first to discover prompt_id.
citepatch_get_visibility_summaryReturns AI visibility metrics for a project window. Requires project_id. Metrics include numerator/denominator style cells from CitePatch Monitor — never treat as causal proof.
citepatch_get_workspaceReturns the organization, accessible projects, granted scopes, and connection identity for this token. Input must be empty — never pass organization_id. Use before other tools to discover project_id values.
citepatch_list_opportunitiesLists playbook cards for a project. Requires project_id. When actionable coverage is present, it links prompt_id -> gap_identity -> playbook_state; missing means no playbook exists. Judgements (impact/confidence/effort) are proposed, not measured.
citepatch_list_promptsLists tracked prompts (and clusters) for a project so you can discover prompt_id values. actionable_gaps links prompt_id -> gap_identity -> playbook_state; playbook_state=missing means no playbook exists for that gap. Requires project_id. Optional cluster_id and status filters. Prefer before citepatch_get_prompt_answers.
citepatch_submit_patchSubmit an externally authored patch for an owned opportunity. First call without confirmed returns confirmation_required + one-use handle; second call with confirmed=true executes. BeforeSHA is verified against the live connector. Consumes MeterPatches; does not debit generation credits.