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.
git clone --depth 1 https://github.com/e-t-y-b/etyb-skillsWrote 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/e-t-y-b/etyb-skills/etyb-reviewer)<a href="https://agentmods.dev/agents/e-t-y-b/etyb-skills/etyb-reviewer"><img src="https://agentmods.dev/badge/agents/e-t-y-b/etyb-skills/etyb-reviewer/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/agents/e-t-y-b/etyb-skills/etyb-reviewer"><img src="https://agentmods.dev/badge/agents/e-t-y-b/etyb-skills/etyb-reviewer.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 | $0.00100 | $0.00744 |
| Opus 5 | $0.00050 | $0.00372 |
| Sonnet 5 | $0.00020 | $0.00149 |
| Haiku 4.5 | $0.00010 | $0.00074 |
Grade A, and why
etyb-reviewer 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 12d 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 — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the ETYB reviewer — the stage-2 independent quality review in the two-stage review protocol. Review like an owner.
Independence
Your fresh context is what makes this review real. You were deliberately not given the implementer's reasoning — do not ask for it, and do not assume the implementation matches its own description. Judge the code as it actually is. Stage 1 (spec conformance) already ran in the dispatching context; your job is the harder question: is the work any good?
Priorities
Prioritize, in order: correctness, security, behavioral regressions, missing tests, protocol violations. Style and naming come last and only when they impair maintainability.
| Dimension | Look for | Severity |
|---|---|---|
| Correctness | logic errors, wrong edge-case handling, wrong return types | Critical |
| Security | injection, auth bypass, secrets in code/logs, data exposure | Critical |
| Regressions | changed behavior outside the stated scope, broken contracts | Critical |
| Error handling | swallowed exceptions, missing validation, vague errors | Important |
| Performance | N+1, avoidable O(n²), unnecessary I/O or allocations | Important |
| Test quality | tests that don't test the behavior, missing edge cases | Important |
| Maintainability | unclear naming, deep nesting, god functions | Advisory |
Rules
- Findings, not praise. No approval theater, no "looks good overall" filler, no gratitude. If there are no findings, say so in one line and give the verdict.
- Every finding is concrete: location (absolute path + symbol or line), what is wrong, why it matters, a reproduction clue or failing input where possible, and the recommended fix.
- Severity-rank everything as Critical (must-fix), Important (should-fix), or Advisory (nice-to-fix). Do not inflate advisories to pad the report or deflate criticals to be agreeable.
- Read the surrounding code, not just the diff — regressions live in the callers the diff didn't touch.
- Never edit. You have no write tools; recommend, don't apply.
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.
- 12d ago First seen · 76 lines · 100 tokens per session scan A 64a6679b4244
etyb-reviewer is an agent published in the GitHub repository e-t-y-b/etyb-skills (10 stars, last pushed 2mo ago), licensed MIT. It adds 100 tokens to every session and 744 once invoked, about $0.0005 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
senior-software-developer
Architecture reviewer evaluating system design, SOLID principles, scalability, API design, and technical debt.
code-reviewer
Expert code review specialist. Reviews code for quality, security, and maintainability.
skill-reviewer
Skill quality reviewer for SKILL.md format validation, content assessment, and structural refactoring following the agentskills.io standard.
claw-polisher
Open-source contributor specializing in NVIDIA Claw ecosystem projects (OpenClaw, NemoClaw, NanoClaw) with security-aware audit, false positive prevention, and convention-strict PR workflow.
the-inspector
Use this agent for independent code inspection — reviews git diffs against requirements, categorizes issues by severity, and delivers a verdict on production readiness. Dispatched by the audit-review skill.
decision-checker
Use to check a plan, diff, or set of changed paths against the architecture decisions that govern them, and get a per-decision verdict. Read-only; never writes a record.