Borrowing it
Nothing to install: this file belongs to jhoblitt/skill-review-claude. 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/jhoblitt/skill-review-claude/main/AGENTS.mdgit clone --depth 1 https://github.com/jhoblitt/skill-review-claudeWrote 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/jhoblitt/skill-review-claude/agents-md)<a href="https://agentmods.dev/instructions/jhoblitt/skill-review-claude/agents-md"><img src="https://agentmods.dev/badge/instructions/jhoblitt/skill-review-claude/agents-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/jhoblitt/skill-review-claude/agents-md"><img src="https://agentmods.dev/badge/instructions/jhoblitt/skill-review-claude/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 | $0.00833 | $0.00833 |
| Opus 5 | $0.00417 | $0.00417 |
| Sonnet 5 | $0.00167 | $0.00167 |
| Haiku 4.5 | $0.00083 | $0.00083 |
Grade A, and why
skill-review-claude AGENTS.md 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 9d 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository conventions
Where the rules live
plugins/skill-review/skills/skill-review/SKILL.md owns the shared review
contract — its bullets are the list, and this register does not keep a second
copy of them. Each of the four axis rules lives in its own reference beside
its corollaries: references/drift.md, references/concurrency.md,
references/token-usage.md, references/security.md.
plugins/skill-review/agents/*.md carry no rules in their bodies. Each names
the two files above that its axis runs under and nothing else, so prose there
that explains an axis is a finding rather than a rendering. Their frontmatter
is the one home for that agent's own tier and tool roster, and its
description renders the axis under the table below.
Rationale stays with the rule it explains, wherever that rule lives.
Sanctioned audience renderings
Seven places re-explain some of that for a different audience instead of
pointing at it. Under references/drift.md's corollary that is a restatement,
so this is a deliberate exception to the one-normative-home rule rather than an
application of it: a marketplace listing cannot cite a contract at a reader who
has not installed it, and a dispatcher deciding whether to load the skill has
not read it either. The exception is bounded by the obligations below,
and each rendering carries only what this table says:
| rendering | may carry |
|---|---|
README.md, intro and axes table |
the axis names, each rule in a line, what each catches, why these failures survive ordinary review, the modes, and the anchoring rule |
README.md, "How a review runs" diagram |
the modes, the axis names, the one-agent-per-axis dispatch, the verify-before-reporting step, the anchoring rule, and the verdict rule, as the shape of one pass |
README.md, "Scope" section |
the scope boundary, the verdict rule, and report-never-fix, stated as outcomes a user sees |
plugins/skill-review/.claude-plugin/plugin.json, description |
the axis names, the four rule names, and the modes |
.claude-plugin/marketplace.json, both description fields |
the axis names and the modes |
plugins/skill-review/skills/skill-review/SKILL.md, frontmatter description |
the axis names, what each catches, the modes, and when the gate mode is entered |
plugins/skill-review/agents/*.md, frontmatter description |
the axis name, its rule in a line, and how the agent is dispatched |
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.
- 9d ago First seen · 62 lines · 833 tokens per session scan A 1ebadd5cc609
skill-review-claude AGENTS.md is an instructions file published in the GitHub repository jhoblitt/skill-review-claude (2 stars, last pushed 25d ago), licensed Apache-2.0. It adds 833 tokens to every session, about $0.0042 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 2026-08-31.
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).
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.