> ## Documentation Index > Fetch the complete documentation index at: https://docs.sixtyfour.ai/llms.txt > Use this file to discover all available pages before exploring further. # MCP Server > Connect Sixtyfour intelligence tools and docs to your AI assistant using the Model Context Protocol. Sixtyfour provides two MCP servers: | Server | URL | Auth | | - | - | - | | `sixtyfour-docs` | `https://docs.sixtyfour.ai/mcp` | None required | | `sixtyfour-intelligence` | `https://mcp.sixtyfour.ai/mcp` | API key or OAuth | ## Setup The `sixtyfour-docs` server requires no authentication. The `sixtyfour-intelligence` server requires either an API key or OAuth for authentication. ### Cursor Install in Cursor Or add to `~/.cursor/mcp.json` (global) or `.cursor/mcp.json` (project) manually. Replace `YOUR_API_KEY` with your key. ```json theme={null} { "mcpServers": { "sixtyfour-docs": { "url": "https://docs.sixtyfour.ai/mcp" }, "sixtyfour-intelligence": { "url": "https://mcp.sixtyfour.ai/mcp?api_key=YOUR_API_KEY" } } } ``` Cursor also supports OAuth for the intelligence server. If you omit the API key, Cursor will authenticate via an OAuth browser flow on first use. Cursor only supports 60 character tool names and descriptions. If you see a warning about tool names being too long in Cursor, the MCP server will still work as Cursor will truncate the tool name and description with a hash suffix. See this [post for more details](https://forum.cursor.com/t/posthog-mcp-tool-names-too-long/155408). ```json theme={null} { "mcpServers": { "sixtyfour-intelligence": { "url": "https://mcp.sixtyfour.ai/mcp" } } } ``` ### VS Code Install in VS Code Or add to `.vscode/mcp.json` in your project root manually. Replace `YOUR_API_KEY` with your key. ```json theme={null} { "servers": { "sixtyfour-docs": { "type": "http", "url": "https://docs.sixtyfour.ai/mcp" }, "sixtyfour-intelligence": { "type": "http", "url": "https://mcp.sixtyfour.ai/mcp?api_key=YOUR_API_KEY" } } } ``` ### Windsurf Add to `~/.codeium/windsurf/mcp_config.json`. Replace `YOUR_API_KEY` with your key. ```json theme={null} { "mcpServers": { "sixtyfour-docs": { "url": "https://docs.sixtyfour.ai/mcp" }, "sixtyfour-intelligence": { "url": "https://mcp.sixtyfour.ai/mcp?api_key=YOUR_API_KEY" } } } ``` After saving the configuration, reload your MCP servers or restart your IDE. ### Claude Code Run these commands in your terminal. Add `--scope user` to make a server available across all projects. **Docs MCP** ```bash theme={null} claude mcp add --transport http sixtyfour-docs "https://docs.sixtyfour.ai/mcp" ``` **Intelligence MCP — OAuth (recommended)** Claude Code will prompt you to authenticate via browser on first use: ```bash theme={null} claude mcp add --transport http sixtyfour-intelligence "https://mcp.sixtyfour.ai/mcp" ``` **Intelligence MCP — API key** ```bash theme={null} claude mcp add --transport http sixtyfour-intelligence "https://mcp.sixtyfour.ai/mcp?api_key=YOUR_API_KEY" ``` ### Claude Desktop In Claude Desktop, add two custom connectors: | Name | Remote MCP server URL | | - | - | | `Sixtyfour Docs` | `https://docs.sixtyfour.ai/mcp` | | `Sixtyfour Intelligence` | `https://mcp.sixtyfour.ai/mcp` | Claude Desktop handles authentication for the intelligence server via OAuth on first connection. ### Codex Install via the [Sixtyfour Codex plugin](https://github.com/sixtyfour-ai/sixtyfour-codex-plugin), or add the servers to `~/.codex/config.toml` manually. #### Plugin install Register the marketplace, then install the plugin. Start a new Codex session after installing — plugins load at session start. **Codex CLI:** ```bash theme={null} codex plugin marketplace add sixtyfour-ai/sixtyfour-codex-plugin codex plugin add sixtyfour@sixtyfour ``` Check installed plugins with `codex plugin list`. To install this plugin in the ChatGPT desktop app instead of the CLI, see [ChatGPT](#chatgpt). Plugins require ChatGPT Work mode or Codex — they aren't available in standard Chat, the IDE extension, or mobile. Work mode currently requires a Pro, Pro Lite, Enterprise, or Edu plan. On Enterprise, Edu, and Business workspaces, an admin must allow installing marketplace plugins under Workspace Settings → Plugins before it appears for members. On first use, authenticate the intelligence server. Run `codex mcp login sixtyfour-intelligence` when prompted. #### Manual config Add the servers to `~/.codex/config.toml`, then restart Codex and sign in when prompted. **Docs MCP:** ```toml theme={null} [mcp_servers.sixtyfour-docs] command = "npx" args = ["-y", "mcp-remote", "https://docs.sixtyfour.ai/mcp"] ``` **Intelligence MCP:** ```toml theme={null} [mcp_servers.sixtyfour-intelligence] command = "npx" args = ["-y", "mcp-remote", "https://mcp.sixtyfour.ai/mcp"] ``` #### Reset a stuck sign-in session The manual `config.toml` + `mcp-remote` method caches its OAuth session in `~/.mcp-auth`, separate from Codex's own session state. If sign-in stops working, clear the cached session for the intelligence server: ```bash theme={null} find ~/.mcp-auth -type f -name "e7e3f47f4a14aca4e636f38d171a9178_*" -delete ``` Restart Codex — it prompts you to sign in again via browser. This only affects the Sixtyfour session; other `mcp-remote`-based connections are untouched. To avoid clearing the cache altogether, switch to the [plugin install](#plugin-install) method, which supports `codex mcp login`/`codex mcp logout` for reconnecting. ### ChatGPT The Sixtyfour plugin is published in the [ChatGPT Plugins Directory](https://chatgpt.com/plugins/plugin_asdk_app_6a514c90130c8191bb9a04df335d5ac6) — the same plugin installed via the CLI marketplace command in [Codex](#codex). View in ChatGPT Plugins 1. Open the listing above, or turn on Work mode in ChatGPT (web or desktop app) and search "Sixtyfour" in Plugins. 2. Select the plugin and install it. 3. Authenticate the intelligence server when prompted. See [Codex](#codex) for plan requirements and workspace admin settings that gate plugin installs. ### Smithery Install the `sixtyfour-intelligence` server through [Smithery](https://smithery.ai/servers/sixtyfour/sixtyfour-intelligence), which handles configuration for Claude Desktop and other supported clients. 1. Open the [Sixtyfour Intelligence listing on Smithery](https://smithery.ai/servers/sixtyfour/sixtyfour-intelligence). 2. Select your client (for example, Claude Desktop) and follow the install command Smithery provides. 3. On first use, authenticate via OAuth or paste your Sixtyfour API key when prompted. *** ## Documentation MCP reference The `sixtyfour-docs` server gives your AI assistant direct access to Sixtyfour API references, guides, and code examples. | Tool | Purpose | | - | - | | `search_sixtyfour_ai_documentation` | Semantic search across Sixtyfour docs | | `query_docs_filesystem_sixtyfour_ai_documentation` | Read specific pages and run `rg`/`cat`/`tree` against the docs filesystem | **Example prompts** * *"How do I use the company-intelligence endpoint?"* * *"What parameters does find-email accept?"* * *"Show me an example of using webhooks with Sixtyfour."* *** ## Intelligence MCP reference The `sixtyfour-intelligence` server exposes the tools below, grouped by category. **Account** | Tool | Purpose | | - | - | | `sixtyfour_check_balance` | Return the organization's current credit balance | **Company search** | Tool | Purpose | | - | - | | `sixtyfour_company_get_searchable_fields` | List all filterable field names for company search | | `sixtyfour_company_explore_field_values` | Get top values for a field (e.g. valid industry names) | | `sixtyfour_company_search` | Search companies with a MongoDB-style filter query | | `sixtyfour_company_next_page` | Fetch the next page of company search results | | `sixtyfour_company_get_company_details` | Expand a result handle into full company data | **People search** | Tool | Purpose | | - | - | | `sixtyfour_people_get_searchable_fields` | List all filterable field names for people search | | `sixtyfour_people_explore_field_values` | Get top values for a field | | `sixtyfour_people_search` | Search people with a MongoDB-style filter query | | `sixtyfour_people_next_page` | Fetch the next page of people search results | | `sixtyfour_people_get_person_details` | Expand a result handle into full person data | Set `count_only` to `true` on `sixtyfour_people_search` to return `total_available` without profiles, result handles, or pagination data. When the backend cannot provide an exact count, `total_available` is `null`: ```json theme={null} { "query": { "locationCountry": "US" }, "count_only": true } ``` **Similar people (lookalike)** | Tool | Purpose | | - | - | | `sixtyfour_find_similar_start` | Find people similar to 1–1000 seed profiles — returns a `task_id` and `search_id` | | `sixtyfour_find_similar_status` | Poll a `sixtyfour_find_similar_start` `task_id` until `completed` | | `sixtyfour_find_similar_results` | Page the ranked matches of a completed search | Pass seeds as `linkedin_urls` (1–1000 LinkedIn profile URLs) or `csv_text` (the full text of a CSV with a LinkedIn URL column, up to 1000 rows), not both. The server infers the traits the seeds share. Use `guidance` only for hard requirements, such as `"only people in the US"`. Seed profiles never appear in the results. A search typically takes 1–3 minutes. Poll with `sixtyfour_find_similar_status`, not `sixtyfour_enrichment_job_status`; set `wait_seconds` (up to 25) to wait server-side between polls. Page the results by `search_id`, then by the returned `next_cursor`. Each row has a 0–100 similarity score, its rank, and the signals that matched. `sixtyfour_find_similar_results` is billed per returned row. `sixtyfour_find_similar_start` starts a billable search; if a call times out, do not retry blindly — check status first. **Async enrichment and contact lookup** | Tool | Purpose | | - | - | | `sixtyfour_company_intelligence_start` | Start a company intelligence job — returns a `task_id` | | `sixtyfour_people_intelligence_start` | Start a people intelligence job — returns a `task_id` | | `sixtyfour_find_email_start` | Find email addresses for 1–100 leads — returns a `task_id` | | `sixtyfour_find_phone_start` | Find phone numbers for 1–100 leads — returns a `task_id` | | `sixtyfour_reverse_email_start` | Resolve the person behind one or more email addresses — returns a `task_id` | | `sixtyfour_reverse_phone_start` | Resolve the person behind one or more phone numbers — returns a `task_id` | | `sixtyfour_enrichment_job_status` | Poll any `_start` tool's `task_id` until `completed`, `failed`, `cancelled`, `terminated`, or `timed_out` | | `sixtyfour_enrich_linkedin` | Enrich a single LinkedIn profile or company URL — returns the result directly | Tools ending in `_start` submit a job and return a `task_id`. Poll `sixtyfour_enrichment_job_status` with that `task_id` until the response shows `completed`. The completed response includes `result`, `charge_amount`, and `task_type`. `sixtyfour_enrich_linkedin` is synchronous — it returns the result directly. Enrichment start tools create billable jobs. If a call times out, do not retry blindly — the job may still have been created. Check status with `sixtyfour_enrichment_job_status` first. **Bulk intelligence** | Tool | Purpose | | - | - | | `sixtyfour_people_bulk_intelligence_start` | Research the same `struct` for many people in one job — returns a `task_id` | | `sixtyfour_company_bulk_intelligence_start` | Research the same `struct` for many companies in one job — returns a `task_id` | Pass the entities as `rows` (a list of identifier objects, such as name, company, and LinkedIn URL) or `csv_text` (CSV text with a header row), not both. `struct` is required: each key becomes an output field, described in plain English or as `{"description": "...", "type": "str"}`. The start response includes `estimated_cost_credits`, a floor estimate. Poll with `sixtyfour_enrichment_job_status`. The completed response includes signed download URLs for the result files under `results`, and `charge_credits`. Inline rows travel in the model's context, so these tools suit up to a few hundred rows assembled from searches. For larger files, use the [web app](https://app.sixtyfour.ai) or the [bulk intelligence API](/api-reference/endpoint/bulk-intelligence) directly. Bulk intelligence start tools create billable jobs. If a call times out, do not retry blindly — check status with `sixtyfour_enrichment_job_status` first. **Age verification** | Tool | Purpose | | - | - | | `sixtyfour_verify_age_start` | Start an age verification job for a person — returns a `task_id` | | `sixtyfour_verify_age_status` | Poll a `sixtyfour_verify_age_start` `task_id` until `completed`, `failed`, `cancelled`, `terminated`, or `timed_out` | `sixtyfour_verify_age_start` and `sixtyfour_verify_age_status` are a dedicated pair — poll with `sixtyfour_verify_age_status`, not `sixtyfour_enrichment_job_status`. The completed response includes `date_of_birth`, `age`, `evidence`, and `sources`. See [Age Verification](/api-reference/endpoint/age-verification) for the full response shape. `sixtyfour_verify_age_start` creates a billable job. If a call times out, do not retry blindly — the job may still have been created. Check status with `sixtyfour_verify_age_status` first. **Feedback** | Tool | Purpose | | - | - | | `send_feedback()` | Report an issue or provide feedback to Sixtyfour | **Example prompts** * *"Search for US SaaS companies with 100–5000 employees."* * *"Research Sixtyfour — return headquarters, headcount, funding stage, and key executives."* * *"Enrich this person: Saarth Shah, CEO & Co-Founder at Sixtyfour."* * *"Find the CTO and VP of Engineering at Sixtyfour."* * *"Find the work email for Saarth Shah at Sixtyfour."* * *"Find a phone number for Jane Doe, VP of Sales at Acme Corp."* * *"Who owns the email [jane@acme.com](mailto:jane@acme.com)?"* * *"Verify the age of Jane Doe, who works at Acme Corp."* * *"Find 50 more people like the ones in this CSV."* * *"Find the work email and GitHub profile for each of these 20 people."* * *"How many Sixtyfour credits do I have left?"* This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.