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 commands/axiomantic/spellbook/code-review-auditgit clone --depth 1 https://github.com/axiomantic/spellbookWrote 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/commands/axiomantic/spellbook/code-review-audit)<a href="https://agentmods.dev/commands/axiomantic/spellbook/code-review-audit"><img src="https://agentmods.dev/badge/commands/axiomantic/spellbook/code-review-audit.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 | $0.00025 | $0.01249 |
| Opus 5 | $0.00013 | $0.00624 |
| Sonnet 5 | $0.00005 | $0.00250 |
| Haiku 4.5 | $0.00003 | $0.00125 |
Grade A, and why
code-review-audit 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 4d 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 — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Code Review: Audit Mode (--audit [scope])
Invariant Principles
- Zero tolerance — surface any deviation; the noise cost is accepted deliberately.
- Build the failure — a check nobody has watched fail is a claim, not a mechanism.
- Evidence Over Assertion — every finding needs a file:line reference.
- Severity Honesty — Critical=security/data loss; Important=correctness; Minor=style.
Scopes
(none)=branch changes, file.py, dir/, security, all
Passes: Correctness > Security > Performance > Maintainability > Edge Cases
Audit Posture — zero tolerance
An audit is a GATE, not a courtesy pass. This posture governs --audit only; the
other modes of code-review are the explicit opt-in to a lighter pass and keep
ordinary judgement about what is worth raising.
- Surface ANY deviation: rule violation, logic bug, design smell, untested behavior, inconsistency — anything off.
- Be adversarial. Verify each finding to filter false positives, but default to flagging when in doubt.
- An audit that "found nothing" on a non-trivial diff is a FAILED audit, not a clean one. Treat an empty finding list as evidence about the audit, not about the code.
Build the failure. Do not just check the author's claim. Both actions cost the same dispatch. Building the failure finds more problems.
- Ask "can I make this fail?" Do not ask "does this pass?" The second question only checks the author's own transcript. The first question does not.
- A check that has never failed is not proven. If nobody has watched a check's failure path fire, the check is a claim, not a mechanism.
- For a claimed clean result — a mutation that should change nothing, a guard that should stay silent — prove the change reached the code under test. A no-op edit looks the same as a correct no-effect. Only the exit status cannot tell them apart.
- Reproduce the defect before you fix it. If you never saw the gap yourself, you are guessing at the gap. A guessed fix cannot be tested.
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.
- 4d ago First seen · 106 lines · 25 tokens per session scan A 25ead4a413ed
code-review-audit is a command published in the GitHub repository axiomantic/spellbook (10 stars, last pushed yesterday), licensed MIT. It adds 25 tokens to every session and 1,249 once invoked, about $0.0001 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 commands, from other repositories
audit-agents-skills
Audit quality of agents, skills, and commands in a Claude Code project.
boldguy-linkedin-answer
Draft a LinkedIn comment on a post, referencing the guide's relevant section, in Flow style with AI markers removed.
security-audit
Comprehensive security audit of your project AND Claude Code configuration. Analyzes secrets exposure, injection surfaces, dependencies, hook security, and produces a scored security posture assessment.
audit-quiz-coverage
Find quiz coverage gaps from recent guide/CHANGELOG/CC-releases changes and propose new questions.
sync
Check guide/landing synchronization status.
worktree
Create a git worktree for isolated feature/fix work.