agentmods.dev

loopgate_harness AGENTS.md

Project instructions for Loopgate Harness, including repository boundaries, forbidden files, quality checks, documentation, and commit rules. A harness is a set of checks that controls how work is validated.

In plain words
What is it for?
Use them when modifying Loopgate Harness, running lint, formatting, type checks, tests, preflight checks, and the final gate.
Why use it?
They keep agents from changing protected parts of the repository and define what must pass before work is considered complete.

Instructions file for CodexOpenCode

20 4d ago C scan 727 tokens original MIT
View the source file rxdt/loopgate_harness
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/rxdt/loopgate_harness/agents-md
Clone the repo
git clone --depth 1 https://github.com/rxdt/loopgate_harness

Made for: Codex, OpenCode.

Per session 727 This file is loaded in full into every session.
When invoked 727 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 $0.00727 $0.00727
Opus 5 $0.00364 $0.00364
Sonnet 5 $0.00145 $0.00145
Haiku 4.5 $0.00073 $0.00073

Measured 3d ago against content hash 4b74cce9ddfb, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

loopgate_harness AGENTS.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 3d 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.

- Never run destructive git commands (`rm -rf`, `git reset --hard`, `git branch -D`) unless the user explicitly asks; verify each risky step.
AGENTS.md · 81 lines

How it starts

The opening of the file, as written. The whole thing — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.

FUNCTIONALITY

Repository boundary

  • Work only inside this repository; create and use scratchpad/ for temporary files.
  • Do not read, write, or search outside the repo unless the user explicitly asks.
  • Do not edit forbidden paths: AGENTS.md, .githooks/, harness/, preferences/ .github/, pyproject.toml. The gate enforces this.
  • Do not make assumptions. Check the code before moving forward.

Quality

  • The quality minimum is code in harness/gate.py (forbidden to agents). Leave it unchallenged.
  • Strengthen tests and coverage. Pass lint, type, and gate checks.
  • Avoid lint suppressions, type-ignores, skipped/xfail tests, or broad exception swallowing.
  • Never run destructive git commands (rm -rf, git reset --hard, git branch -D) unless the user explicitly asks; verify each risky step.
  • Never bypass or reconfigure git hooks.

Commit and gate

  • ralph harness preflight runs on commit => fast lint + format check + plus loop containment.
  • Run harness preflight often, run harness gate before pushing
  • Done means: no forbidden path touched; ralph harness gate is green; your chosen spec and docs/PROJECT_STATUS.md reflect what was built.

Documentation

  • Every agent-maintained .md (docs/PROJECT_STATUS, docs/specs/) stays under 100 lines — distill for the next agent.

Session handoff

  • At <=40% of your context window, stop expanding scope.
    • Update docs/PROJECT_STATUS.md with new state, checks, commit/branch, blockers, and next steps.
    • Update your spec.
    • Leave the repo resumable.
    • Commit through the gate, push, then merge when safe.

RULES

1. Think First

Verify assumptions. Surface confusion. Note tradeoffs.

Before implementing:

  • State assumptions explicitly. If uncertain, ask.
  • If multiple interpretations exist, present them - don't pick silently.
  • If a simpler approach exists, say so. Push back when warranted.
  • If something is unclear, stop. Request clarity.
  • Review source code and tests.

Read the full file on GitHub · 81 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. 3d ago First seen · 81 lines · 727 tokens per session scan C 4b74cce9ddfb

Subscribe to this mod's changes

loopgate_harness AGENTS.md is an instructions file published in the GitHub repository rxdt/loopgate_harness (20 stars, last pushed 4d ago), licensed MIT. It adds 727 tokens to every session, about $0.0036 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 2026-08-30.

CommitsGit and workflowagentagentic-aiautomationclaude-codecodex-cliframework
Related

Other instructions, from other repositories

claude-007-agents CLAUDE.md

Instructions for avivl/claude-007-agents, covering critical: commit message requirements, pre-commit checklist, claude code - advanced ai agent system, 🎯 evil corp motivation framework and agent categories.

avivl/claude-007-agents · 6,131 tokens

ii-agent AGENTS.md

Instructions for Intelligent-Internet/ii-agent, covering ii-agent contributor guide, quick start, repository map, mandatory rules and architecture.

Intelligent-Internet/ii-agent · 3,294 tokens

thinkrail AGENTS.md

Instructions for JetBrains/thinkrail, covering thinkrail, module structure & boundaries (top-priority requirement), engine: pi only, in-process, architecture (three rings) and repo layout.

JetBrains/thinkrail · 4,503 tokens

sortie copilot-instructions.md

Instructions for sortie-ai/sortie, covering sortie coding & review standards, 1. layered imports (downward only; violation is critical), 2. concurrency safety, 3. workspace path safety (critical: security boundary) and 4. persistence (sqlite).

sortie-ai/sortie · 1,030 tokens

copperhead CLAUDE.md

Instructions for copperheadhq/copperhead, covering claude.md, what this repo is, sources of truth, workflow (openspec) and architecture (per spec.md §2).

copperheadhq/copperhead · 1,242 tokens

acodex AGENTS.md

Instructions for maksimzayats/acodex, a project described as: Local MCP automation bridge for the Codex desktop app, with CLI, HTTP server, and live codexapp. tool access.

maksimzayats/acodex · 400 tokens