Borrowing it
Nothing to install: this file belongs to bharat3645/The-Ideal-Harness. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/bharat3645/The-Ideal-Harness/main/CLAUDE.mdgit clone --depth 1 https://github.com/bharat3645/The-Ideal-HarnessWrote 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.
[](https://agentmods.dev/instructions/bharat3645/the-ideal-harness/claude-md)<a href="https://agentmods.dev/instructions/bharat3645/the-ideal-harness/claude-md"><img src="https://agentmods.dev/badge/instructions/bharat3645/the-ideal-harness/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/bharat3645/the-ideal-harness/claude-md"><img src="https://agentmods.dev/badge/instructions/bharat3645/the-ideal-harness/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.03800 | $0.03800 |
| Opus 5 | $0.01900 | $0.01900 |
| Sonnet 5 | $0.00760 | $0.00760 |
| Haiku 4.5 | $0.00380 | $0.00380 |
Grade C, and why
The-Ideal-Harness CLAUDE.md scanned grade C with 2 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 10d 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.
(`rm -rf ~//`, `mkfs`, `dd …of=/dev/`, fork bomb). Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- **Ask:** all `Bash`, `Edit`, `Write`, `WebFetch`; `curl`/`wget`/`nc`; `git push`. How it starts
The opening of the file, as written. The whole thing — 156 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
The Ideal Harness — Project Instructions
Scope rule (overrides global workflow): In this repository, route work through the Ideal Harness's own modules only. Do not invoke the global user-level workflow plugins here — no gsd, caveman, gstack, claude-mem, ruflo, codex-gate, brainstorming, or graphify. This project is the harness; it dogfoods itself. Use its modules, skills, and floor — nothing else.
This repo builds The Ideal Harness: a control-plane around a stateless model. It dogfoods its own
enforcement floor and bootstrap skill via .claude/settings.json. The six modules below are the
only lanes in play here — the harness ships as a single Claude Code plugin, but each module is its
own MCP server/CLI and its own source boundary; treat them as separate lanes, not as a monolith.
The harness modules (the only lanes)
| Need | Module | How to reach it |
|---|---|---|
| Token pressure / large tool output | compress |
Automatic tool_result compression; call ccr_retrieve when you see a <<ccr:HASH>> marker. |
| "What calls X", "where is Y", past decisions | memory |
query_graph (code structure) / memory_search (episodic) instead of re-reading whole files. A durable, project-level architecture decision goes in decisions.md (human-reviewed file), not a tool call. |
| Stale/uncertain library or package knowledge | web |
web_docs (live npm registry metadata/README) / web_fetch (any URL) — policy-gated exactly like the native WebFetch tool. |
| Multi-step build / plan / review | orchestrate |
Brainstorm (no code until approved) → plan → scout locates → plan-critic (different model tier, non-trivial plans only — the dual-model consensus gate) → fresh-context implementer per task → reviewer gate → fix loop. ledger_verify actually re-runs a task's verify.command instead of trusting a self-report. The four agents ship in agents/ (symlinked into .claude/agents/ for dogfood discovery). |
| Any tool call | guard |
Deterministic floor below the model. vet_skill / vet_skill_deep scan a third-party skill (text or a whole directory, the latter adding semgrep/osv-scanner when present) before you trust it. If a call is denied, it is denied for a reason — do not route around it. |
| Substrate (loader, validation, templating) | core |
pnpm validate; skill templating + multi-host generation. |
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.
- 10d ago First seen · 156 lines · 3,800 tokens per session scan C 1135f6287622
The-Ideal-Harness CLAUDE.md is an instructions file published in the GitHub repository bharat3645/The-Ideal-Harness (5 stars, last pushed today), licensed MIT. It adds 3,800 tokens to every session, about $0.0190 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.