mcp_examples

mcp_examples is an agent for coding agents from momentmaker/kaijutsu. It costs 0 tokens per session (1,412 once invoked), scanned A, original, MIT.

Example configurations for connecting deterministic analysis tools to an agent swarm through MCP, a standard way for software tools to communicate with AI agents.

In plain words
What is it for?
Use it as a reference for connecting tools such as Semgrep, a code scanner, and running a shared security-audit analysis.
Why use it?
It lets fixed-rule analyzers return structured findings alongside language-model participants in the same review.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/momentmaker/kaijutsu/mcp_examples
Clone the repo
git clone --depth 1 https://github.com/momentmaker/kaijutsu

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for mcp_examples

README.md
[![agentmods](https://agentmods.dev/badge/agents/momentmaker/kaijutsu/mcp_examples.svg)](https://agentmods.dev/agents/momentmaker/kaijutsu/mcp_examples)
Your own site
<a href="https://agentmods.dev/agents/momentmaker/kaijutsu/mcp_examples"><img src="https://agentmods.dev/badge/agents/momentmaker/kaijutsu/mcp_examples.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,412 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.01412
Opus 5 $0.00000 $0.00706
Sonnet 5 $0.00000 $0.00282
Haiku 4.5 $0.00000 $0.00141

Measured 4d ago against content hash 28d225ce7be1, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

mcp_examples scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

cli/internal/agents/mcp_examples.md · 145 lines

How it starts

The opening of the file, as written. The whole thing — 145 lines — stays where its author put it; the contents beside it link to each section on GitHub.

MCP-as-peer reference integrations

The mcp driver lets deterministic analyzers participate in jutsu swarm alongside LLM peers. Per spec D7 the server exposes a single agreed-upon tool — analyze (overridable via tool_name) — that takes the preset name + severity vocabulary + input body, and returns a JSON array of findings.

This document is illustrative — none of the configs below are run as part of the v0.6 test suite. The mcp driver itself is fully tested against the in-tree stub at testdata/stub_mcp_server/.

semgrep-mcp (security analyzer)

semgrep is a static analysis engine with ~2000 community-maintained rules covering OWASP basics, framework-specific patterns, and common bug classes. A semgrep-mcp wrapper exposing the analyze contract is straightforward to build (or fork from existing community efforts).

# ~/.kaijutsu/agents.yaml
providers:
  semgrep-mcp:
    driver: mcp
    transport: stdio
    command: npx
    args: ["semgrep-mcp"]
    tool_name: analyze
    timeout_seconds: 120
# <repo>/.kaijutsu/agents.yaml
enabled: [claude, codex, antigravity, semgrep-mcp]
$ jutsu swarm security-audit --personas \
    paranoid-security-claude,pragmatic-codex,default-antigravity,default-semgrep-mcp

LLM peers + semgrep run in parallel. Findings tagged [deterministic] in the synthesizer's disagreement table. When semgrep agrees with two of three LLMs at issue+ severity, that's near-certain real signal.

Free-tier note: semgrep community rules cover most OWASP Top-10 patterns for free, no account required. Pro rules (interprocedural taint, secrets, supply-chain CVE matching) require a paid AppSec Platform account but overlap heavily with what the LLM peers catch — non-auth is the right default for jutsu's use case. See README.

eslint-mcp (JavaScript linter)

providers:
  eslint-mcp:
    driver: mcp
    transport: stdio
    command: npx
    args: ["eslint-mcp", "--config", ".eslintrc.json"]
    tool_name: analyze

Read the full file on GitHub · 145 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 4d ago First seen · 145 lines · 0 tokens per session scan A 28d225ce7be1

Subscribe to this mod's changes

mcp_examples is an agent published in the GitHub repository momentmaker/kaijutsu (3 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,412 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.