Vibocode Mcp

io.github.vnbochkarev-netizenPublished by GitHub account vnbochkarev-netizenv0.1.3Updated Nov 7, 2026

Two MCP tools instead of an API catalog: search endpoints, run Python in a sandbox.

VerifiedSTDIODesktop onlyKnowledge & MemoryDeveloper Tools

Overview

AI-generated overview

Local MCP server that replaces a large API tool catalog with two tools: search endpoints and run sandboxed Python that calls allowlisted APIs and keeps memory.

What you can do
  • Search a service's action catalog for endpoints and stored memory facts, returning compact signatures
  • Run model-written Python in a sandbox that calls allowlisted HTTP endpoints through api.get
  • Add, find and read statistics on persistent memory facts across sessions
  • Generate a catalog from an OpenAPI spec instead of writing it by hand
When to use it
Useful when an assistant must work with a large API (GitHub, Stripe, Slack and similar) without spending tens of thousands of tokens on one tool per endpoint, or when you want facts learned in one session to be recalled in the next.
Requirements
Runs locally via npx; needs Node 18+ and Python 3.10+. Requires a catalog.json path in VIBOCODE_CATALOG describing base URL, allowed domains and actions; VIBO_MEM_FILE is optional and enables persistent memory. Network access is limited to the catalog's allowed domains.
Before you install
The sandbox is Python-level (AST checks, resource limits, separate process), not a hardened isolate, so it guards against careless code rather than deliberate attacks. The run tool executes model-written code and can make HTTP requests to the allowlisted hosts, and memory writes persist to the file given in VIBO_MEM_FILE. Session-based endpoints may need a cookie or service token.

Installation

In SourceWeft

  1. Open Vibocode Mcp in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

vibocode-mcp

Two tools instead of an API catalog. An MCP server that keeps the agent's context small and its reach long.

search  — finds the right endpoints and memory facts, returns compact signaturesrun     — executes model-written Python in a sandbox: api.get(path) + vi.add/find/stats

The whole toolset is 217 tokens (o200k_base tokenizer). The same GitHub REST catalog — 1,272 endpoints — costs 79,525 tokens as plain MCP tools: a 366x difference. Slack's 174 endpoints: 42x. Measurements and reproduction commands are in Benchmarks.

Install

bash
npx vibocode-mcp

Requires Node 18+ and Python 3.10+ (the server is Python).

MCP client configuration:

json
{  "mcpServers": {    "vibocode": {      "command": "npx",      "args": ["-y", "vibocode-mcp"],      "env": {        "VIBOCODE_CATALOG": "/path/to/catalog.json",        "VIBO_MEM_FILE": "/path/to/memory.web"      }    }  }}

Action catalog

catalog.json describes a service: base URL, allowed domains and the list of actions.

json
{  "service": "example",  "base_url": "http://127.0.0.1:4174",  "allow_domains": ["127.0.0.1"],  "actions": [    {"name": "health", "method": "GET", "path": "/api/health", "auth": false,     "desc": "Service health: storage, notification channels, server time"}  ]}

The catalog is not written by hand: for OpenAPI it is generated.

bash
python3 tools/openapi_to_catalog.py openapi.json -o catalog.json \  --base-url https://api.example.com --allow api.example.com --service example

What it looks like in use

python
h = api.get("/api/health")print(h["status"], h["data"]["storage"])vi.add("Service health", "health ok, storage local-json")print(vi.stats())

One run call: an API request, a memory write and a stats read. Without memory it is just a sandbox; the value is that the next session does not rediscover what is already known.

What is inside

bin/vibocode-mcp.js        — npm entry point (starts the Python server)vibocode/server.py         — MCP server (JSON-RPC over stdio)vibocode/catalog.py        — catalog search, compact signaturesvibocode/aliases.json      — query alias data (natural language -> catalog terms)vibocode/bridge.py         — bridge: ViBo memory + allowlisted HTTPvibocode/sandbox.py        — AST validation + isolated process with limitsvibocode/runner.py         — model code executortools/openapi_to_catalog.py — catalog generator from OpenAPItests/test_mcp.py          — 19 checksskill/                     — packaging as an agent skill (SKILL.md + CLI)

Security

  • imports: json, math, re, datetime, statistics, itertools
  • forbidden: open, eval, exec, compile, __import__, getattr, setattr, dunder attributes, global/nonlocal, infinite loops
  • network: only api.get(path), only hosts from allow_domains
  • limits: 40 s wall clock, 1.5 GB memory, 20 s CPU, output capped at 4,000 characters
  • writing the same fact to memory twice is skipped

This is a Python-level sandbox (AST + rlimits + a separate process), not a V8 isolate: it protects against accidental and careless code, not against a deliberate OS-level attack.

Tests

bash
python3 tests/test_mcp.py     # protocol, search, memory, API, limits, blocks

Benchmarks

Tokenizer: o200k_base (tiktoken). Catalogs are generated from official OpenAPI specs. "Full catalog" means one tool per endpoint, with name, description and an empty input schema — the lower bound, generous to the alternative.

CatalogActionsFull catalogvibocodeRatio
GitHub REST1,27279,525 tokens217366x
Stripe61238,020 tokens217175x
Slack Web API1749,129 tokens21742x

For GitHub this is 40% of a 200,000-token window before the user says a word.

End-to-end: one run call with 196 tokens of code performed three requests to api.github.com, wrote a fact to memory and read the stats back. The whole five-step exchange: 2,202 tokens.

Search quality

Search is lexical and alias-aware. aliases.json maps the words a user types to catalog terms, so a query can be phrased naturally rather than in endpoint vocabulary.

Shipped configuration (English aliases):

CatalogQueriesTop-1Top-4
Stripe1283%92%
GitHub REST540%80%

With a language pack loaded through VIBOCODE_ALIASES (a mixed 28-query set, half of it non-English): Stripe 80% / 95%, GitHub 38% / 75%.

Known weakness in both cases: queries that describe a whole scenario instead of an object. Next step: hybrid BM25 + embeddings.

Limitations

  • transport is line-delimited JSON-RPC (not Content-Length); sufficient for stdio clients
  • session-based endpoints need a cookie or a service token
  • the catalog is static: new endpoints are added to catalog.json (or generated from OpenAPI)
  • search is lexical, without a vector index

License

MIT. ViBo memory is a separate product with its own license.

Source: README.md at commit 5ed68c5

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v0.1.3LatestNov 7, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts