Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add fakoli/fakoli-plugins/plugin install fakoli-plugin-criticWrote 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/skills/fakoli/fakoli-plugins/plugin-review)<a href="https://agentmods.dev/skills/fakoli/fakoli-plugins/plugin-review"><img src="https://agentmods.dev/badge/skills/fakoli/fakoli-plugins/plugin-review/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/skills/fakoli/fakoli-plugins/plugin-review"><img src="https://agentmods.dev/badge/skills/fakoli/fakoli-plugins/plugin-review.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 | $0.00046 | $0.00717 |
| Opus 5.5 | $0.00018 | $0.00287 |
| Sonnet 5.5 | $0.00009 | $0.00143 |
| Haiku 4.5 | $0.00005 | $0.00072 |
Grade A, and why
plugin-review 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 1mo 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.
What it actually says
Plugin Review
Identify the target host and actual plugin root before applying any checklist. A Markdown role file is not automatically a registered Codex agent type. Inspect the current tool/agent inventory and inherit the current model settings. Perform focused review locally when delegation is unavailable; when parallel agents are available and authorized, assign independent read-only scopes.
Read the specialist reference for each changed surface:
| Surface | Reference |
|---|---|
| Manifests, catalog paths, versions | Structure critic |
| Skill triggers and executable workflows | Skill critic |
| Hook events, paths, failure behavior | Hook critic |
| MCP wiring, transport and secret handling | MCP critic |
| Host-specific agent configuration | Agent critic |
These legacy checklists include Claude conventions and project preferences. Apply the target host's current contract rather than treating every preference as a universal requirement:
- Codex uses
.codex-plugin/plugin.json; keep skills under a declared/discoverable skill directory. Do not add unsupportedagentsorcommandskeys just because those folders exist. - Resolve marketplace local sources from the marketplace root, and manifest paths from the plugin root. Read actual resolved files. Preserve other source types, policies and ordering.
- Hooks are supported in current Codex, but event coverage, trust and payloads depend on the host.
${PLUGIN_ROOT}is the native root variable;${CLAUDE_PLUGIN_ROOT}is compatible in documented Codex hook contexts. Verify the actual context before reporting a portability defect. - Core Agent Skills metadata requires a valid name and meaningful description. Three examples, minimum 250-character descriptions, mandatory hard gates and specific model tiers are local preferences unless the repository explicitly requires them.
agents/openai.yamlcan contain invocation policy without an interface block. Runtime tool names and transports come from the exposed schema; do not require invented MCP prefixes or pre-allows.- Skill references should load on demand. Read the files needed to validate the changed behavior; do not demand eager loading of the entire bundle to conduct a small scoped review.
Trace each suspected defect from trigger to effect, cite the relevant file/line and applicable
contract, and reproduce it safely when possible. Do not run hook commands, send messages, install
plugins, or invoke remote mutating tools merely to audit their declarations. Tests may use temp
fixtures or mocked servers. Mark unreproduced concerns SUSPECTED and explain the missing evidence.
Return only actionable findings with severity, trigger, effect, evidence and a concrete fix, followed by checks performed and limitations. Separate MUST FIX correctness/contract failures from SHOULD FIX maintenance issues and optional advice. A clean review is a valid result. Critics report; edits are made only when the user also requested fixes.
Current primary references: OpenAI plugin packaging, Agent Skills, and the target host's current hook/MCP docs.
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.
- 1mo ago First seen · 53 lines · 46 tokens per session scan A f24c29c5a894
plugin-review is a skill published in the GitHub repository fakoli/fakoli-plugins (4 stars, last pushed 27d ago), licensed MIT. It adds 46 tokens to every session and 717 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-09-06.
Other skills, from other repositories
audit
Read-only quality gate that audits already-built work against its stated requirement and Definition of Done before it is allowed to move to 2done/. Produces a severity-ranked decision proposal (BLOCKER/MAJOR/MINOR/NIT) with evidence; the judgement never edits, only a recorded post-audit step may fix small findings.…
pr-vetting
Use when the user wants to thoroughly investigate, vet, review, or security-scan a pull request before deciding how to handle it - especially PRs from external or unknown contributors. Runs a parallel multi-agent investigation (technical, security, value/fit, contributor reputation) and merges the findings into one…
generic-code-reviewer
Review code for bugs, security vulnerabilities, performance issues, accessibility gaps, and CLAUDE.md workflow compliance. Supports any tech stack - HTML/CSS/JS, React, TypeScript, Node.js, Python, NestJS, Next.js, and more. Use when completing features, before commits, or reviewing pull requests.
verify-findings
Verify code-review or security-review findings for false positives using deep codebase tracing, framework-aware analysis, and web research. Produces a .verified.md report alongside the original. Use when a review report has been generated and needs independent verification, or when user runs /verify-findings, mentions…
santa-method
Send a finished change to two reviewers who did not write it, using one rubric, and stop after three rounds. Use before a public page, a priced artifact, or a pull request that changes behavior.
diag
Read-only root-cause diagnosis for a symptom, bug, failure, or unexpected behavior. Establishes the mechanism at file:line and reads up the facts instead of guessing, before any fix is attempted. Produces a diagnosis report; the fix is always a separate step gated by an explicit GO. Use when something is broken…