Borrowing it
Nothing to install: this file belongs to MichelKerkmeester/skilled-agent-harness_spec-driven-loops. 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/MichelKerkmeester/skilled-agent-harness_spec-driven-loops/main/AGENTS.mdgit clone --depth 1 https://github.com/MichelKerkmeester/skilled-agent-harness_spec-driven-loopsWrote 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/michelkerkmeester/skilled-agent-harness_spec-driven-loops/agents-md)<a href="https://agentmods.dev/instructions/michelkerkmeester/skilled-agent-harness_spec-driven-loops/agents-md"><img src="https://agentmods.dev/badge/instructions/michelkerkmeester/skilled-agent-harness_spec-driven-loops/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80ร15 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/michelkerkmeester/skilled-agent-harness_spec-driven-loops/agents-md"><img src="https://agentmods.dev/badge/instructions/michelkerkmeester/skilled-agent-harness_spec-driven-loops/agents-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 | $0.10828 | $0.10828 |
| Opus 5 | $0.05414 | $0.05414 |
| Sonnet 5 | $0.02166 | $0.02166 |
| Haiku 4.5 | $0.01083 | $0.01083 |
Grade A, and why
skilled-agent-harness_spec-driven-loops AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories โ prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency โ measured 2d 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.
How it starts
The opening of the file, as written. The whole thing โ 507 lines โ stays where its author put it; the contents beside it link to each section on GitHub.
AI Assistant Framework (Universal Template)
Universal behavior framework defining guardrails, standards, and decision protocols.
Multi-Repository Architecture
Universal Framework: Code work routes through the sk-code skill, which auto-detects the active surface and loads its patterns and verification; unrecognized surfaces trigger a disambiguation question. Detection markers and per-surface patterns live in .opencode/skills/sk-code/SKILL.md ยง2 Smart Routing.
Repo-Local Layer: This document is shared across repositories, so anything belonging to ONE repository lives beside it in that repository's root REPO RULES.md โ a router to per-rule documents under repo-rules/. Gate 5 (ยง2) makes reading it mandatory before your first write, where the repository has one. Its rules bind exactly as this document's do; where the two appear to disagree, this document wins.
The Iron Law: NO completion claims without running stack-appropriate verification.
1. ๐จ CRITICAL RULES โ HARD BLOCKERS
The Four Laws โ HARD BLOCKERS (cannot be overridden)
Expanded by
scope-discipline.md(Law 2),evidence-and-proof.md(Law 3), androot-cause-and-debugging.md(Law 4).
- READ FIRST โ Never edit a file without reading it first. Understand context before modifying.
- SCOPE LOCK โ Only modify files explicitly in scope. NO "cleaning up" or "improving" adjacent code. Scope in
spec.mdis FROZEN. - VERIFY โ Syntax checks and tests MUST pass before claiming completion. NO blind commits.
- HALT โ Stop immediately if uncertain, if line numbers don't match, or if tests fail.
Law 4 blocks forward progress and completion while a check is failing. A failing check may enter the bounded remediation loop in Section 3, but the hard stop remains until the authoritative gate passes.
PLAN-WORKFLOW LOCK โ HARD BLOCKER (cannot be overridden)
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.
- 2d ago Changed ยท +21 tokens per session 168bd435053f
- 3d ago Changed ยท +187 tokens per session 0530c3cd7c99
- 5d ago First seen ยท 507 lines ยท 10,620 tokens per session scan A e7bc9461fe41
skilled-agent-harness_spec-driven-loops AGENTS.md is an instructions file published in the GitHub repository MichelKerkmeester/skilled-agent-harness_spec-driven-loops (35 stars, last pushed today), licensed MIT. It adds 10,828 tokens to every session, about $0.0541 per session on Opus 5. 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 2026-09-03.
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).
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.
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.