Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/heavy3-ai/code-auditnpx agentmods add skills/heavy3-ai/code-audit/skillWrote 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/heavy3-ai/code-audit/skill)<a href="https://agentmods.dev/skills/heavy3-ai/code-audit/skill"><img src="https://agentmods.dev/badge/skills/heavy3-ai/code-audit/skill/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/heavy3-ai/code-audit/skill"><img src="https://agentmods.dev/badge/skills/heavy3-ai/code-audit/skill.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 4 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 25 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 50 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Rogue Agent · line 87 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
- medium Rogue Agent · line 413 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
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 | $0.00022 | $0.07696 |
| Opus 5.5 | $0.00009 | $0.03078 |
| Sonnet 5.5 | $0.00004 | $0.01539 |
| Haiku 4.5 | $0.00002 | $0.00770 |
Grade A, and why
h3 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 16d 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 — 798 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Heavy3 Code Audit - The Multi-Model Code Review for Coding Agents
Sponsored by Heavy3.ai - Multi-model AI for high-stakes decisions
You are helping the user get AI-powered code reviews via OpenRouter.
All features are free and open source:
- Single model review with DeepSeek V4 Pro (strong reasoning at low cost)
- 3-model council with GPT 6.1 Sol + Gemini 4 Argon + Grok 4.7
- Web search integration
- Up to 200K token context
Arguments
$ARGUMENTS can contain:
Explicit targets (no confirmation needed):
pr <number>- Review a GitHub pull request by numberplan <path>- Review a specific plan file<file>.md- Shorthand for plan review (any .md file)<range>- Review a commit range (e.g.,HEAD~3..HEAD,abc123..def456)
Scope modifiers:
--staged- Force review of only staged changes--commit- Force review of the last commit only
Mode options:
--council- Use 3-model council (GPT 6.1 Sol + Gemini 4 Argon + Grok 4.7)--free- Use rotating free model from config--model <name>- Override model (shortcuts: glm, gpt, kimi, deepseek, free)
Smart Detection
When /h3 is invoked without explicit targets, automatically detect intent and confirm with user.
Detection Priority
| Priority | Condition | Action |
|---|---|---|
| 1 | Explicit argument provided | Execute directly, no confirmation |
| 2 | Uncommitted changes exist | Confirm: review changes? |
| 3 | No changes + plan detected | Confirm: review the plan? |
| 4 | No changes + no plan | Ask: review commits or specify target? |
Step-by-Step Smart Detection Workflow
Step 1: Check for explicit arguments
If $ARGUMENTS contains any of these, skip detection and execute directly:
pr <number>→ PR reviewplan <path>→ Plan review of specific file<file>.md(any markdown file path) → Plan review<range>(commit range likeHEAD~3..HEAD) → Code review of range--staged→ Staged changes review--commit→ Last commit review
What ships with it
13 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- config.json 1.4 KB
- requirements-dev.txt 255 B
- scripts/council.py 23 KB runs code
- scripts/list-free-models.py 5.3 KB runs code
- scripts/review.py 29 KB runs code
- scripts/tests/__init__.py 33 B runs code
- scripts/tests/conftest.py 12 KB runs code
- scripts/tests/test_context.py 22 KB runs code
- scripts/tests/test_council.py 17 KB runs code
- scripts/tests/test_free_models.py 11 KB runs code
- scripts/tests/test_review.py 17 KB runs code
- scripts/tests/test_streaming.py 13 KB runs code
- TEST_PLAN.md 8.5 KB
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.
- 16d ago Changed · +1 lines d6dcd3cfce04
- 1mo ago Changed 5a9c3db0ad57
- 1mo ago Changed 24a86a1c4fca
- 2mo ago First seen · 797 lines · 22 tokens per session scan A 6057ddebb1d2
h3 is a skill published in the GitHub repository heavy3-ai/code-audit (46 stars, last pushed 16d ago), licensed MIT. It adds 22 tokens to every session and 7,696 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
autoreview
Structured code review through the shared OpenClaw agent-skills installation.
omh-code-review
This is an OMH code-review workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
explore-unknowns
Guide the user through a quadrant walk that maps the unknowns of a task — open by listing the known knowns, then work through known unknowns, unknown knowns, and unknown unknowns one stage at a time, ending with a complete four-quadrant map in the user's hands. Use when a request is ambiguous or underspecified, the…
re0-merge
Review and land an external contribution the way this suite does: gate it against the thesis, land it with the author's credit intact, complete a new skill rather than merging it raw, then approve, credit, and explain before closing. Use when reviewing a pull request, as any collaborator or maintainer, not only the…
asplos-writing-style
Use when drafting or revising ASPLOS prose — engineering the first two pages that rapid review actually reads, stating the cross-layer insight as one quotable sentence, structuring per-layer mechanism sections, writing bounded claims, and fitting everything into 11 self-contained pages that include figures and…