Code-Fundi MCP Server
OfficialProvides tools for indexing and managing GitHub repositories within the Code-Fundi codebase map, enabling semantic search, repository intelligence, and blast-radius analysis.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Code-Fundi MCP Server Map the codebase and analyze the blast radius of changing the auth service."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Code-Fundi MCP Server
A production-grade Model Context Protocol (MCP) server for the Code-Fundi API: a codebase map and blast-radius engine for AI coding agents. Enables any MCP-compatible AI assistant (Claude, Claude Code, Cursor, Windsurf, VS Code Copilot, and more) to search, research, index, and scan code repositories through natural language, and see what a change breaks before it ships.
Built with FastMCP (TypeScript) and Zod for schema validation.
Getting started for AI agents
If you are an AI agent (or configuring one) and no CODEFUNDI_API_KEY is set, you must self-authenticate before using search, repos, chat, or other authenticated tools. The auth tools (code-fundi-auth-*) work without a pre-existing key.
Which path to use
Situation | Action |
| Use any tool immediately β the server sends it as |
No API key configured | Run the OTP self-auth flow below (recommended) or password sign-in |
Zero-config MCP setup (no API key)
You can start the MCP server without CODEFUNDI_API_KEY and authenticate at runtime:
{
"mcpServers": {
"code-fundi": {
"command": "npx",
"args": ["-y", "@codefundi/code-fundi-mcp"]
}
}
}Self-authenticate with OTP (recommended)
Code-Fundi uses Supabase-backed auth (POST /v2/auth/authenticate, /v2/auth/verify, /v2/auth/resend). OTP emails contain a 6-digit code β magic links are not supported on this path.
Ask the human user for their email address.
Call
code-fundi-auth-authenticatewith:auth_mode:"otp"email: the user's emailshould_create_user:truefor a new account,falsefor a returning user
Tell the user to check their inbox for a 6-digit code. The API may return
verification_required: trueandapi_key.key_state: "agent_pending"until verification completes.Ask the user for the 6-digit OTP (human-in-the-loop β you cannot guess or bypass this step).
Call
code-fundi-auth-verifywith the sameemailand thetoken(6 digits).On success, the MCP server automatically configures the API key in memory for all subsequent tool calls in this session.
If the code expired or was not received, call code-fundi-auth-resend with the same email, then repeat step 5.
Example dialogue:
Agent: What email should I use to sign in to Code-Fundi?
User: dev@example.com
Agent: [calls code-fundi-auth-authenticate] I've sent a 6-digit code to dev@example.com. Please paste it here.
User: 482913
Agent: [calls code-fundi-auth-verify] You're signed in. I can now search and index your repositories.Password sign-in (alternative)
For existing accounts with a password, call code-fundi-auth-authenticate with auth_mode: "password", the user's email, should_create_user: false, and the password parameter. The MCP client sends the password only in the X-CodeFundi-Auth-Password header (never in the JSON body). Production requires HTTPS. Returning users may receive an active API key immediately without a separate verify step.
After authentication
The API key is held in memory for the lifetime of the MCP server process. It is not persisted across IDE or MCP restarts.
Recommend the user add the key to their MCP config as
CODEFUNDI_API_KEYso future sessions start authenticated.A FREE-tier account and API key are created automatically on first signup.
Errors
HTTP status | What to do |
401 Unauthorized | No valid key β run the OTP flow above, or set |
429 Too many requests | Auth endpoints are rate-limited per IP; wait for |
Related MCP server: Codebase Intelligence MCP Server
Features
Every tool below is backed by the same codebase map: structural dependencies, call graph, and blast radius, indexed once and queried in milliseconds.
π Semantic & grep code search across your indexed codebase map
π§ AI-powered research: search plus AI analysis in a single call
π¦ Repository management: index, status, README, listing, public catalog
π°οΈ Repository intelligence: cross-repo dependency map, blueprint, Blast-Radius Guard (impact analysis before you merge)
π File documentation: AI-generated docs for any indexed file
π Usage statistics: query usage, activity, language breakdowns
π Agent-driven authentication: sign up/sign in via OTP without pre-configured keys
π¬ AI chat & model insight: direct conversation with Code-Fundi AI, model catalog, and per-tier limits
Quick Start
Install from npm (recommended)
Install the package (includes a pre-built dist/). The code-fundi-mcp binary is on your PATH when installed globally, or available via npx without cloning the repo:
npm install -g @codefundi/code-fundi-mcpOr add it to a project:
npm install @codefundi/code-fundi-mcpInstall from source (this repository)
git clone https://github.com/Code-Fundi/code-fundi-mcp.git
cd code-fundi-mcp
npm install
npm run buildConfigure
Option A β API key (fastest): set your key as an environment variable:
export CODEFUNDI_API_KEY=your_api_key_hereOption B β no API key: skip the env var and let the agent self-authenticate at runtime. See Getting started for AI agents.
Zero-config MCP example (no env block):
{
"mcpServers": {
"code-fundi": {
"command": "npx",
"args": ["-y", "@codefundi/code-fundi-mcp"]
}
}
}Use with Claude Desktop
After a global install (npm i -g @codefundi/code-fundi-mcp), point MCP at the published binary (no path to dist/index.js required):
{
"mcpServers": {
"code-fundi": {
"command": "@codefundi/code-fundi-mcp",
"env": {
"CODEFUNDI_API_KEY": "your_api_key_here"
}
}
}
}If the binary is not on your PATH, use npx (downloads or uses the local package and runs the same entrypoint):
{
"mcpServers": {
"code-fundi": {
"command": "npx",
"args": ["-y", "@codefundi/code-fundi-mcp"],
"env": {
"CODEFUNDI_API_KEY": "your_api_key_here"
}
}
}
}Use with Cursor
Same pattern as Claude: command plus optional args only, no manual path to the repo:
{
"mcpServers": {
"code-fundi": {
"command": "npx",
"args": ["-y", "@codefundi/code-fundi-mcp"],
"env": {
"CODEFUNDI_API_KEY": "your_api_key_here"
}
}
}
}When developing inside this repo, you can run npm run dev or npx tsx src/index.ts without building first.
Development Mode
npm run dev # Run with tsx (no build needed)
npx fastmcp inspect src/index.ts # Open MCP Inspector UI
npx fastmcp dev src/index.ts # Test with MCP CLITools Reference (27 tools)
Covers the Code-Fundi V2 API for codebase mapping and blast-radius analysis: search (including search-with-chat / research), repositories (list, index, status, readme, public catalog), repository intelligence (map, blueprint, radius), files, history, statistics, API keys, authentication, plus Fundi chat (POST /v1/fundi/chat) and the V2 model catalog / limits (GET /v2/models, GET /v2/models/limits).
Search
Tool | Description |
| Semantic and grep search across your indexed codebase map, with filters |
| Search plus AI-synthesized analysis of matching code |
Repositories
Tool | Description |
| List indexed repositories with pagination |
| Index a new GitHub repository into your codebase map |
| Check repository indexing status |
| Get repository README documentation (deprecated, prefer blueprint) |
| Browse the global catalog of indexed public repositories (no key needed) |
Repository Intelligence
Tool | Description |
| Cross-repository dependency map: how services and packages actually connect |
| README plus dependency and convention overview (successor to repo-readme) |
| Blast-Radius Guard: every file and function that breaks before you merge (PRO+) |
Files
Tool | Description |
| List files in a repository |
| Get AI-generated file documentation |
History
Tool | Description |
| List query history with filters |
| Get full details of a history entry |
| Get conversation thread messages |
Statistics
Tool | Description |
| Query-type usage breakdown |
| Daily activity statistics |
| Programming language usage |
Authentication
Tool | Description |
| Start OTP/password auth flow (no key needed) |
| Verify OTP code and obtain API key |
| Resend OTP verification email |
| List API keys (masked) |
| Regenerate API key |
| Disable an API key by ID ( |
Chat & Models
Tool | Description |
| Fundi AI chat ( |
| List the curated chat model catalog ( |
| Get AI model limits and tier configuration ( |
Authentication
Two modes: pre-configured API key (CODEFUNDI_API_KEY in MCP config) or agent-driven OTP/password auth at runtime. Full step-by-step instructions, tool names, and error handling are in Getting started for AI agents at the top of this README.
Environment Variables
Variable | Required | Default | Description |
| No* | β | API key for authentication |
| No |
| API base URL override |
* Required unless using agent-driven auth tools.
License
MIT
Available Tools
27 toolscode-fundi-activity-statsARead-only
Get daily activity statistics for a given date range. Shows queries per day with credit costs.
Inputs: range.
code-fundi-auth-authenticateA
Start Code-Fundi authentication (POST /v2/auth/authenticate). No existing API key required. Use auth_mode otp for new or returning users: set should_create_user true on first signup, false on sign-in. Sends a 6-digit email OTP (not a magic link); then call code-fundi-auth-verify after the human user provides the code. Password mode: pass password here (sent as X-CodeFundi-Auth-Password header only); should_create_user false for sign-in. On success with an active key, the API key is configured in-memory for this MCP session only.
Inputs: email (required), auth_mode (required), password, should_create_user.
code-fundi-auth-resendA
Resend the OTP email (POST /v2/auth/resend) when the previous code expired or was not received. Use after code-fundi-auth-authenticate, before code-fundi-auth-verify. Default type is signup for new accounts.
Inputs: email (required), type.
code-fundi-auth-verifyA
Complete OTP sign-in (POST /v2/auth/verify). Requires the 6-digit code from the user's email after code-fundi-auth-authenticate. The human user must supply the token β ask them explicitly. On success, activates the API key and configures it in-memory for all subsequent tools in this MCP session. Persist CODEFUNDI_API_KEY in MCP config if the user wants auth to survive restarts.
Inputs: email (required), token (required).
code-fundi-chatARead-only
Send a message to Code-Fundi AI (Fundi chat: POST /v1/fundi/chat). Supports threading, optional code context, indexed repository knowledge (knowledge_id), embeddings memory, and voice mode. Responses are streamed by the API and returned as plain text (or JSON when the server uses JSON mode).
Inputs: prompt (required), embed, model, voice, +4 more.
code-fundi-conversationARead-only
Get all messages in a conversation thread by its conversation ID.
Inputs: conversation_id (required).
code-fundi-disable-api-keyADestructive
Disable a Code-Fundi API key by UUID (DELETE /v2/keys/{key_id}). Use code-fundi-list-api-keys to find key IDs. Does not rotate the active key unless you disable the one in use.
Inputs: key_id (required).
code-fundi-file-docsARead-only
Get AI-generated documentation for a specific file in an indexed repository. Returns detailed documentation including summaries, functions, dependencies, and code analysis.
Inputs: file_id (required), repo_key (required), fields.
code-fundi-history-itemARead-only
Get the full details of a single history entry by its ID, including the complete response and knowledge sources.
Inputs: history_id (required).
code-fundi-index-repoA
Index a new GitHub repository in Code-Fundi, or update an existing one. After indexing, the repository's files will be searchable via code-fundi-search.
Inputs: url (required), branch, update.
code-fundi-language-statsARead-only
Get programming language usage statistics across all indexed repositories.
Inputs: no named parameters.
code-fundi-list-api-keysARead-only
List all Code-Fundi API keys (masked) for the authenticated account.
Inputs: no named parameters.
code-fundi-list-filesARead-only
List files in an indexed Code-Fundi repository. Supports search by file name/path and pagination. Use the returned file IDs with code-fundi-file-docs to get documentation.
Inputs: repo_key (required), limit, order, offset, +2 more.
code-fundi-list-historyARead-only
List your Code-Fundi query history with optional filters for date range, category, endpoint, and repository.
Inputs: to, from, limit, offset, +6 more.
code-fundi-list-modelsARead-only
List the curated Code-Fundi chat model catalog (GET /v2/models) with providers, required tier, and context length.
Inputs: no named parameters.
code-fundi-list-public-reposARead-only
Browse the global catalog of indexed public repositories (across all Code-Fundi users). Supports pagination, an updated_since cursor, and optional embedding of each repo's largest files. This endpoint does not require an API key.
Inputs: limit, since, offset, file_count, +1 more.
code-fundi-list-reposARead-only
List all indexed repositories in your Code-Fundi account. Supports pagination, scope filtering (private/public), and name search.
Inputs: limit, order, scope, offset, +2 more.
code-fundi-model-limitsARead-only
Get the AI model limits and tier configuration for your account (GET /v2/models/limits): subscription tokens, the active model's context/knowledge limits, and account limits such as max repositories, files per repo, history retention, and organization access.
Inputs: no named parameters.
code-fundi-regenerate-api-keyADestructive
Regenerate the Code-Fundi API key. The full key is shown only once. The new key is automatically configured for all subsequent tool calls.
Inputs: no named parameters.
code-fundi-repo-blueprintARead-only
Get the canonical repository intelligence overview: README documentation, top dependencies, top functions/variables, and aggregated coding conventions (PRO+ only). This is the successor to code-fundi-repo-readme. Consumes credits.
Inputs: repo_key (required), demo, conventions_path_prefix.
code-fundi-repo-mapARead-only
Build a cross-repository dependency map showing where a dependency appears across the source repo, optional compared repos, and (optionally) the public index. Returns structured JSON. Consumes credits.
Inputs: repo_key (required), demo, type, limit, +4 more.
code-fundi-repo-radiusARead-only
Run blast-radius / file impact analysis for one or more files or dependencies. Returns projected data flows, entry points, and call edges. Requires PRO, ENTERPRISE, or ADMIN tier (STARTUP is not eligible) and consumes credits.
Inputs: repo_key (required), demo, file, files, +4 more.
code-fundi-repo-readmeARead-only
Get the parsed README documentation for a repository. Provide either a repository UUID or a clone URL. Deprecated in favor of code-fundi-repo-blueprint, which returns the same README plus dependency and convention intelligence.
Inputs: repo_key (required).
code-fundi-repo-statusBRead-only
Check the indexing status of a repository by its clone URL.
Inputs: url (required).
code-fundi-researchARead-only
Search Code-Fundi repositories AND get an AI-synthesized analysis of the results. This performs a search, then streams an AI response that analyzes the matching code files. Use this when you need both code context and an intelligent summary/explanation.
Inputs: query (required), model, scope, fields, +12 more.
code-fundi-searchARead-only
Search across indexed Code-Fundi repositories using semantic search, grep over docs, or grep over code. Returns matching files with similarity scores, paths, and optional documentation. Use scan_mode to choose between semantic (vector), grep_docs (substring on documentation), or grep_code (substring on source code).
Inputs: query (required), scope, fields, repo_ids, +11 more.
code-fundi-usage-statsARead-only
Get query-type usage statistics for a given date range. Shows breakdown by query type with counts, credit costs, and average durations.
Inputs: range.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelogβ¦
TDQS
Scored across 27 tools
Most tools have clearly distinct purposes, but a few pairs could cause confusion: code-fundi-search vs. code-fundi-research (research wraps search with AI analysis), code-fundi-usage-stats vs. code-fundi-activity-stats, and the deprecated code-fundi-repo-readme overlapping with code-fundi-repo-blueprint. Descriptions are detailed enough to disambiguate, but the overlaps are noticeable.
All tools share the 'code-fundi-' prefix, but the remainder follows inconsistent patterns: some use verb_noun (list-repos, index-repo, list-files), others are noun phrases (repo-map, repo-blueprint, file-docs), and some are noun_verb (auth-authenticate, auth-verify). This mix of conventions makes it harder to predict tool names.
At 27 tools, the server exceeds the 25-tool threshold for 'too many'. The count is inflated by a deprecated tool (code-fundi-repo-readme) and a set of six auth-related tools that could be consolidated. While the broad scope partly justifies the number, the presence of redundant/deprecated tools makes it feel overbuilt.
The tool surface covers the core Code-Fundi workflows: repository indexing and listing, search/research, file and blueprint retrieval, stats, history, auth, API key management, chat, and model limits. Missing operations include repository removal (de-indexing) and more granular repo configuration, but these are minor gaps for the primary use cases.
Maintenance
Related MCP Connectors
A MCP server built for developers enabling Git based project management with project and personalβ¦
An MCP server that gives your AI access to the source code and docs of all public github repos
Repository knowledge graph MCP server for codebase understanding and debugging.
MCP server for building and testing AI agents with multi-model experimentation and insights.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceAn MCP server that gives AI agents structured code understanding and precise code intelligence via local indexing of AST, call graphs, and semantic search.77 npm4Apache 2.0
- AlicenseNot gradedqualityDmaintenanceA production-grade MCP server providing advanced code intelligence capabilities such as semantic search, dependency analysis, and natural language Q&A across multiple programming languages.MIT
- AlicenseNot gradedqualityBmaintenanceMCP server for local-first code intelligence, providing structural code graph, semantic search, and impact analysis to AI agents.2MIT
- AlicenseNot gradedqualityBmaintenanceA production-ready MCP server that enables AI assistants to intelligently understand, analyze, edit, navigate, and review software projects with multi-workspace support, Git integration, and semantic search.1MIT