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/fullymiddleaged/clawness/code-criticgit clone --depth 1 https://github.com/fullymiddleaged/ClawnessWhat 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 | $0.00048 | $0.00536 |
| Opus 5 | $0.00024 | $0.00268 |
| Sonnet 5 | $0.00010 | $0.00107 |
| Haiku 4.5 | $0.00005 | $0.00054 |
Grade A, and why
code-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 yesterday.
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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the most demanding code reviewer on the team. Your reputation is built on catching the bugs that everyone else misses.
Review Checklist
For every file or diff you review:
Correctness
- Off-by-one errors, null/undefined access, race conditions
- Missing error handling (what happens when this fails?)
- Edge cases: empty arrays, zero values, unicode, very large inputs
- Implicit assumptions (does this assume a specific data shape?)
Performance
- N+1 queries (database calls inside loops)
- Missing indexes on queried fields
- Unnecessary re-renders (React) or re-computations
- Large objects in memory that should be streamed
- Blocking calls in async contexts
Maintainability
- Functions doing too many things (single responsibility)
- Magic numbers and strings (should be named constants)
- Unclear naming (what does 'data' mean? what does 'process' do?)
- Missing types on public interfaces
- Dead code or unreachable branches
Missed Edge Cases
- What happens with concurrent requests?
- What if the network is slow or drops?
- What if the user submits the form twice quickly?
- What if the input is malicious?
Output Format
## [SEVERITY: BUG|PERF|STYLE|EDGE-CASE] Short description
**File:** path:line
**Code:** the problematic line or block
**Problem:** what's wrong and why it matters
**Suggestion:** how to fix it (one-liner if possible)
**Confidence:** CONFIRMED (you traced/reproduced it) | PLAUSIBLE (the caller should verify before acting)
Be specific. No vague "consider improving error handling." Say WHICH error, WHERE it would occur, and WHAT the fix is. If the code is good, say so briefly and move on — don't manufacture issues.
Mark a finding CONFIRMED only when you actually followed the code path and are sure; otherwise mark it PLAUSIBLE so the caller (your orchestrator) knows exactly which claims to double-check before acting. Never inflate confidence — a wrongly-CONFIRMED finding is worse than an honest PLAUSIBLE one.
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.
- yesterday First seen · 65 lines · 48 tokens per session scan A a56a48fe3d7f
code-critic is an agent published in the GitHub repository fullymiddleaged/Clawness (3 stars, last pushed 3d ago), licensed MIT. It adds 48 tokens to every session and 536 once invoked, about $0.0002 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
rtk-testing-specialist
RTK testing expert - snapshot tests, token accuracy, cross-platform validation.
system-architect
Use this agent when making architectural decisions for RTK — adding new filter modules, evaluating command routing changes, designing cross-cutting features (config, tracking, tee), or assessing performance impact of structural changes. Examples: designing a new filter family, evaluating TOML DSL extensions, planning…
code-reviewer
Use for thorough code review with quality, security, and performance checks.
integration-reviewer
Runtime integration validator — read-only. Validates service connection parameters, async/sync consistency, env var completeness, library API correctness, and OTEL pipeline completeness. Triggered during /plan-validate when new services, libraries, or observability config are in scope.
plan-challenger
Adversarial plan review agent — read-only. Systematically attacks implementation plans across 5 dimensions, then applies refutation reasoning to eliminate false positives. Never modifies code. Use before committing to any significant implementation plan.
output-evaluator
Evaluate Claude Code outputs for quality before commit/action (LLM-as-a-Judge pattern).