MCP.Market

Agentsbestfriend MCP server

by TheRealFloatDev·io.github.TheRealFloatDev/agentsbestfriend·v0.12.0
No reviews yet

MCP that provides fast, token-efficient tools for navigating, searching, and understanding codebases

B74/100grade B
Write a reviewInstall
What users say
No reviews yet
Be the first
Safety scan
B74/100

full report

Adoption
New

0 stars52 downloads/wk

Reviews

Write one

Nobody has reviewed Agentsbestfriend yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

Agentsbestfriend tools (13)

write = sends, deletes, buys or posts

Read from the package source without running it. The installed server may list more.

  • abf_chunk

    Smart file chunking by symbol boundaries. Returns actual source code. Use EXACTLY ONE of these modes: - symbol: pass a symbol name to get its full source code directly - chunk_index: pass a 0-based chunk index to get that chunk's code - (neither): returns a chunk overview listing \u2014 use this first to discover available chunks, then call again with chunk_index to retrieve code

  • abf_context_bundle

    Bundle multi-file context around an entry point in a single call. Traverses the import graph from the entry file, collecting symbols and optionally source code from all connected files. Replaces multiple read_file + abf_symbols + abf_dependencies calls with one compact result. Modes: - "signatures": compact \u2014 shows exported symbols with type signatures for all files (lowest tokens) - "full":

  • abf_conventions

    Detect codebase conventions by analyzing file names, folder structure, design patterns, and config files. Returns detected conventions with confidence scores and examples. Purely heuristic \u2014 no LLM required. Useful to understand a project's style before making changes.

  • abf_dependencies

    Show imports and reverse dependencies (imported_by) for a file.

  • abf_file_summary

    Search across LLM-generated file summaries using full-text search. Returns files whose summaries match the query, ranked by relevance (BM25). Useful when exploring a codebase by concept rather than exact code text. Default mode is "or" \u2014 matches files containing ANY of the query terms, ranked by how many match. Use "and" mode to require ALL terms to be present. Requires summaries to be genera

  • abf_git

    Query git history, blame, and diff for the project.

  • abf_impact

    Find all files and lines that reference a given symbol name. Useful for change impact analysis.

  • abf_index

    Manage the file index: check status, trigger rebuild, incremental update, or generate LLM summaries. Rebuild/update will auto-generate summaries when Ollama is available.

  • abf_notes

    Agent notepad \u2014 persist notes across sessions in a local SQLite database (.abf/notes.db). Supports save, get, list, search (full-text via FTS5), update, and delete actions. Use this to store task context, decisions, TODOs, or anything that should survive between conversations. Notes are project-scoped and independent from the code index (survive re-indexing).

  • abf_ping

    Health check tool. Returns server status and project root. Use this to verify the ABF MCP server is running correctly.

  • abf_project_overview

    Get a comprehensive overview of the current project. Returns detected tech stack, frameworks, entry points, directory structure with purposes, language distribution, config files, and architectural patterns. Very token-efficient way to orient yourself in a new codebase. No index required.

  • abf_search

    Search code in the project using multiple modes: - "exact": Fast ripgrep-based search for exact strings or regex patterns. Returns matching lines with file path, line number, and surrounding context. - "keyword": Exploratory search \u2014 splits the query into keywords, scores every file by keyword density, returns the top matching files ranked by relevance. Best when you're not sure of exact name

  • abf_symbols

    Get the symbol outline (functions, classes, methods, types) of a file.

Public scan report

scanner v0.1.9 · 2026-09-27 · same rubric, same numbers if you re-run it

1 medium
  • Code scan2 source files scanned20/25
  • –Live reliabilityno gateway calls yet and no remote to proben/a
  • –Tool poisoningtools not inspected (local package is not executed); not countedn/a
  • Auth qualitylocal package, no credentials required12/15
  • Maintenancelast push 158 days ago8/15
  • Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10

Findings (1)

  • mediumeval / new Function usedexec.eval
    dist/index.js: …ode, sch); const makeValidate = new Function(`${names_1.default.self}`, `${names_1.de…
Overall 74/100. Components that don't apply are left out of the denominator. Any critical finding is an F.RubricAppeal a findingJSON

How Agentsbestfriend compares

Against every graded MCP server, from our own scans and uptime checks. Updated after each scan run.

Safety score
74/100
higher than 40% of 48,255 servers
Tools
13
the typical server has 7
Tools that change things
0%
typical: 0%. Fewer write tools means less an agent can break
Downloads a week
52
more than 43% of 12,191 servers with a package

What its tools let an agent do: 13 read

Install Agentsbestfriend in Claude Code, Cursor or VS Code

Runs npx -y agentsbestfriend on your machine. Read the scan report first; the gateway never runs local packages.

claude mcp add agentsbestfriend -- npx -y agentsbestfriend
Add to Cursor

What the publisher says

From the Agentsbestfriend repository's README, as published. We do not edit it. Read it on GitHub

AgentsBestFriend (abf)

Give your AI coding agents superpowers — a local MCP server for fast, token-efficient code navigation, search & analysis.

Works with VS Code Copilot, Cursor, Claude Code/Desktop, Codex, Cline, Zed, Gemini CLI, Goose, OpenCode, and any other MCP-compatible agent.

Why?

AI coding agents waste tokens re-reading files and searching blindly. ABF gives them purpose-built tools that return exactly what they need — in compact, structured responses that preserve context.

Tools

Install

npm install -g agentsbestfriend

Prerequisites

  • Node.js ≥ 20
  • ripgrep — brew install ripgrep (macOS) / apt install ripgrep (Linux)
  • git
  • Ollama (optional) — for summaries & semantic search: ollama.com

Quick Start

abf init

abf init walks you through everything:

  1. Indexes your project — discovers and indexes all files via git ls-files
  2. Generates LLM summaries & embeddings (if Ollama is running) — with live (12/80) progress
  3. Adds .abf/ to .gitignore — prompts before writing
  4. Installs ABF as an MCP server for your agents — you pick which agents (Cursor, VS Code, Claude Code, etc.) and whether to use npx agentsbestfriend start (always latest) or abf start (local install)

Manual Agent Setup

If you prefer to configure manually, add ABF as a stdio MCP server. Using npx is recommended — it always runs the latest published version without requiring a global install:

VS Code / GitHub Copilot (.vscode/mcp.json):

{
  "servers": {
    "abf": {
      "command": "npx",
      "args": ["agentsbestfriend", "start"]
    }
  }
}

Shortened. The full README is on GitHub.

Nothing above is checked by us. What we check is on the safety report.

Agentsbestfriend: common questions

Is Agentsbestfriend MCP server safe?
Mostly: it is graded B (74/100). Read the Agentsbestfriend safety report
How do I install Agentsbestfriend?
It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
Does Agentsbestfriend need an API key?
No secret keys are declared. It reads 1 setting from the environment.
Is Agentsbestfriend maintained?
The last commit was 196 days ago (2026-04-22). The latest release is v0.12.0.
What can I use instead of Agentsbestfriend?
Servers from other publishers that do the same job: Codebase Memory MCP server.
How does Agentsbestfriend compare with other MCP servers?
safety score 74/100 (higher than 40% of 48,255 servers); tools 13 (the typical server has 7); tools that change things 0% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.

Alternatives to Agentsbestfriend

Same job from other publishers: the closest match first, then the best rated.

  • Codebase Memory
    Codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.
    C