joint-chiefs CLAUDE.md

joint-chiefs CLAUDE.md is an instructions file for coding agents from djfunboy/joint-chiefs. It costs 1,702 tokens per session, scanned C, original, MIT.

Repository instructions for Joint Chiefs, a multi-model AI code-review tool with command-line, MCP server, macOS app, and key-management interfaces. They describe its architecture, rules, stack, and configuration.

In plain words
What is it for?
Working on Joint Chiefs code, including its Swift packages, AI-provider integrations, command-line tools, MCP server, macOS app, and tests.
Why use it?
They give the coding agent the project context and constraints needed to make compatible changes in the repository.

Instructions file

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 instructions/djfunboy/joint-chiefs/claude-md
Clone the repo
git clone --depth 1 https://github.com/djfunboy/joint-chiefs

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 joint-chiefs CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/djfunboy/joint-chiefs/claude-md.svg)](https://agentmods.dev/instructions/djfunboy/joint-chiefs/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/djfunboy/joint-chiefs/claude-md"><img src="https://agentmods.dev/badge/instructions/djfunboy/joint-chiefs/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 1,702 This file is loaded in full into every session.
When invoked 1,702 The same file — it is already loaded in full.
Security scan C 1 finding. 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 $0.01702 $0.01702
Opus 5 $0.00851 $0.00851
Sonnet 5 $0.00340 $0.00340
Haiku 4.5 $0.00170 $0.00170

Measured 4d ago against content hash 5407fc27a8bf, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

joint-chiefs CLAUDE.md scanned grade C with 1 finding 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.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

4. **Cold-machine smoke test** — `rm -rf "/Applications/Joint Chiefs.app"`, mount fresh DMG, drag-install, launch, confirm first-run window appears. Signing/notarization verdicts alone don't catch dyld/rpath failures.
CLAUDE.md · 69 lines

How it starts

The opening of the file, as written. The whole thing — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Joint Chiefs

Canonical instruction file. AGENTS.md is a symlink to this file.

Goal: Public MIT-licensed open-source tool. Success = adoption and acquisition value, not revenue. Reliability beats features — prefer minimal correct changes. Don't propose monetization, premium tiers, or scale infrastructure unless asked.

Multi-model AI code review orchestrator. Four surfaces (CLI, stdio MCP server, macOS setup app, keygetter binary) built on one JointChiefsCore engine. Hub-and-spoke debate with Claude as moderator; streams consensus summary. Grounded in MAD research.

Website: https://jointchiefs.ai/ (source: private djfunboy/joint-chiefs-website repo — this repo is the app) Latest release: v0.5.13 — added claude-opus-5 (newest Opus flagship) to the Anthropic picker; claude-fable-5 remains the Anthropic/moderator default. 90+ tests. CFBundleVersion 1777000011. Next session: read the most recent tasks/SESSION-HANDOFF-*.md (gitignored; local-only).

Stack & Architecture

  • Build system: Swift Package Manager. Minimum target macOS 15.0 (Sequoia). Apple Silicon only — no Intel support.
  • Packages: JointChiefsCore (models/providers/orchestrator/credential store/APIKeyResolver), JointChiefsCLI (jointchiefs), JointChiefsMCP (jointchiefs-mcp, stdio), JointChiefsSetup (SwiftUI setup app), JointChiefsKeygetter (jointchiefs-keygetter)
  • MCP dependency: modelcontextprotocol/swift-sdk pinned exact 0.12.0
  • State management: @Observable macro + @State/@Environment/@Bindable. Never mix with ObservableObject.
  • Service pattern: @Environment injection — no singletons.
  • Design system: Agentdeck (monospace-as-identity, warm-charcoal). All colors/fonts/spacing/radii in JointChiefsSetup MUST come from Agentdeck* token files in Sources/JointChiefsSetup/DesignSystem/. Read docs/reference/DESIGN-SYSTEM.md before any UI change. Never hardcode hex or CGFloat.
  • Providers: OpenAI, Anthropic Claude, Google Gemini, xAI Grok, Ollama, OpenAI-compatible (LM Studio/Jan/llama.cpp-server/Msty/LocalAI) — all via REST + SSE streaming
  • Storage: StrategyConfig JSON → ~/Library/Application Support/Joint Chiefs/strategy.json (0600); API keys → credentials.json (same dir, 0600) via keygetter only

Read the full file on GitHub · 69 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 · 69 lines · 1,702 tokens per session scan C 5407fc27a8bf

Subscribe to this mod's changes

joint-chiefs CLAUDE.md is an instructions file published in the GitHub repository djfunboy/joint-chiefs (0 stars, last pushed 24d ago), licensed MIT. It adds 1,702 tokens to every session, about $0.0085 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other instructions, from other repositories

apfel CLAUDE.md

Instructions for Arthur-Ficial/apfel, covering apfel - project instructions, the golden goal, core product (this is what apfel is), byproducts (useful, but not the pitch) and readme.md structure rule.

Arthur-Ficial/apfel · 7,076 tokens

blockrun-mcp AGENTS.md

AGENTS.md instructions for BlockRunAI/blockrun-mcp, covering blockrun mcp, commands, project structure, key dependencies and install in codex.

BlockRunAI/blockrun-mcp · 282 tokens

headroom-desktop CLAUDE.md

Claude Code instructions for gglucass/headroom-desktop, covering claude.md - headroom-desktop, testing rules, wheel bump rules, compression / cache change rules and persistence rules.

gglucass/headroom-desktop · 2,180 tokens

GhidrAssistMCP CLAUDE.md

Claude Code instructions for symgraph/GhidrAssistMCP, covering claude.md, build and development commands, building the extension, set ghidrainstalldir environment variable first and build the extension.

symgraph/GhidrAssistMCP · 1,019 tokens

openrouter-mcp-multimodal AGENTS.md

AGENTS.md instructions for stabgan/openrouter-mcp-multimodal, covering agent instructions, before you ship, releasing (read this before publishing), short version and version files (must all match package.json).

stabgan/openrouter-mcp-multimodal · 793 tokens

intervals-icu-mcp CLAUDE.md

Claude Code instructions for hhopke/intervals-icu-mcp, covering claude.md, project overview, development commands, architecture (quick reference) and tool categories.

hhopke/intervals-icu-mcp · 1,873 tokens