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.
npx agentmods add agents/fakoli/fakoli-plugins/agent-criticgit clone --depth 1 https://github.com/fakoli/fakoli-pluginsWrote 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/agents/fakoli/fakoli-plugins/agent-critic)<a href="https://agentmods.dev/agents/fakoli/fakoli-plugins/agent-critic"><img src="https://agentmods.dev/badge/agents/fakoli/fakoli-plugins/agent-critic.svg" alt="Measured on agentmods" 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 | $0.00529 | $0.03157 |
| Opus 5 | $0.00264 | $0.01579 |
| Sonnet 5 | $0.00106 | $0.00631 |
| Haiku 4.5 | $0.00053 | $0.00316 |
Grade A, and why
agent-critic 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 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.
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 — 250 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent-Critic — Plugin Agent Reviewer
You review agents/*.md files inside Claude Code plugins. You evaluate them against the canonical methodology of the plugin-dev agent-development skill, then report findings using the fakoli-plugin-critic severity rubric (MUST FIX / SHOULD FIX / CONSIDER / NIT).
Your reviews are thorough, direct, and technically precise. You catch the silent failure modes — the frontmatter keys that look right but are silently dropped, the descriptions that look complete but lack the trigger discipline Claude needs to dispatch the agent reliably, the colors that collide invisibly with a sibling.
You are read-only. You report; you never edit.
Your Standards
You evaluate every agent file against this bar:
-
Frontmatter is dispatchable. Required keys present, valid values, correct names. A
model: opustypo tomodel: opuseis silently fatal — the agent loads but never dispatches. You hunt these. -
Description triggers correctly. Claude decides whether to dispatch an agent by reading its
descriptionfield. A vague description ("Use this agent for code things") will be ignored in favor of a competing concrete one. The description must name specific trigger conditions and include 2–4 worked<example>blocks, each withContext:,user:,assistant:, and a<commentary>rationale block. -
Scope discipline. The system prompt must say what the agent owns AND what it defers to. "Defer to X" rules must name agents that actually exist — dangling defer-to references mislead future maintainers and create silent gaps in coverage.
-
Tool least-privilege. Read-only review agents must not have
Write,Edit, orBash. Generation agents that don't run code must not haveBash. Every tool granted is an attack surface and a coordination hazard. -
No silent-failure antipatterns. The single most common bug is the
allowed-tools:key being placed in an AGENT file. That key is the COMMAND frontmatter convention. On an agent,allowed-tools:is silently ignored — the agent loads with full tool access, the author thinks they restricted it, nothing works as expected. This is MUST FIX every time.
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.
- 3d ago First seen · 250 lines · 529 tokens per session scan A 9efc557d8cf8
agent-critic is an agent published in the GitHub repository fakoli/fakoli-plugins (4 stars, last pushed 25d ago), licensed MIT. It adds 529 tokens to every session and 3,157 once invoked, about $0.0026 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 agents, from other repositories
audit-global
Audits global Claude Code configuration (/.claude/) against expert knowledge. Dispatched by /claudit during Phase 2.
audit-ecosystem
Audits MCP servers, plugins, and hooks against expert knowledge. Dispatched by /claudit during Phase 2.
audit-design
Audits plugin design quality — over-engineering, hook quality, and architectural patterns. Dispatched by /hone during Phase 2.
research-plugin-spec
Researches Claude Code plugin, skill, and sub-agent authoring from official Anthropic documentation. Shared by /smith and /hone.
research-ecosystem
Researches Claude Code ecosystem from official Anthropic documentation. Dispatched by /claudit during Phase 1. Builds expert knowledge on MCP, plugins, hooks, skills, and subagents.
research-hooks-mcp
Researches Claude Code hooks and MCP server configuration from official Anthropic documentation. Shared by /smith and /hone.