advanced-code-review-verify

advanced-code-review-verify is a command for coding agents from axiomantic/spellbook. It costs 19 tokens per session (3,926 once invoked), scanned A, original, MIT.

A code-review verification step that checks reported findings against the actual code and removes findings that cannot be confirmed. A pull request, or PR, is a proposed set of code changes for review.

In plain words
What is it for?
Use it to verify review findings at specific lines, discard refuted issues, and mark findings that remain inconclusive.
Why use it?
It reduces false alarms and makes uncertainty visible. For PR reviews, it uses only the PR's changed code when that is the applicable review source.

Command

Install

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.

agentmods
npx agentmods add commands/axiomantic/spellbook/advanced-code-review-verify
Clone the repo
git clone --depth 1 https://github.com/axiomantic/spellbook

Wrote 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.

agentmods badge for advanced-code-review-verify

README.md
[![agentmods](https://agentmods.dev/badge/commands/axiomantic/spellbook/advanced-code-review-verify.svg)](https://agentmods.dev/commands/axiomantic/spellbook/advanced-code-review-verify)
Your own site
<a href="https://agentmods.dev/commands/axiomantic/spellbook/advanced-code-review-verify"><img src="https://agentmods.dev/badge/commands/axiomantic/spellbook/advanced-code-review-verify.svg" alt="Measured on agentmods" height="20"></a>
Per session 19 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,926 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00019 $0.03926
Opus 5 $0.00010 $0.01963
Sonnet 5 $0.00004 $0.00785
Haiku 4.5 $0.00002 $0.00393

Measured 5d ago against content hash 6cabee8a21c2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

advanced-code-review-verify 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 5d 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.

commands/advanced-code-review-verify.md · 433 lines

How it starts

The opening of the file, as written. The whole thing — 433 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Phase 4: Verification

Invariant Principles

  1. Every finding must be verifiable against actual code: If a finding cannot be verified by reading the file at the specified line, it is not a valid finding.
  2. REFUTED findings must be removed, not just flagged: False positives erode trust. Remove them from final output entirely; log in audit for transparency.
  3. INCONCLUSIVE findings must be clearly marked: Uncertainty is acceptable; hidden uncertainty is not. Mark findings that could not be verified so humans can assess.
  4. PR mode = diff-only source: When reviewing a PR (not a local branch), the diff is the only authoritative code. Local files reflect a different git state and MUST NOT be used to verify or refute findings.

4.0 Pre-Flight: Branch Safety Check

The reviewing-prs skill is the single source for that decision. Load it and apply its review_source decision table; do not re-derive the rule here. This command previously carried its own copy, and the copy had already diverged — it missed the worktree case, so a review running inside a worktree checked out to the PR branch was downgraded to DIFF_ONLY and lost every verification it could have made.

Take review_source from review-manifest.json if Phase 1 recorded it; otherwise compute it per reviewing-prs and write it back to the manifest so the later phases and this one cannot disagree.

Phase 4's consequence of review_source == "DIFF_ONLY":

  • ALL verify_* functions return "INCONCLUSIVE" immediately — do NOT read local files
  • Mark the finding [NEEDS VERIFICATION] in the report
  • Reason: "PR review — local HEAD does not match PR HEAD SHA; local files reflect different code"

Read the full file on GitHub · 433 lines

Changes

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.

  1. 5d ago First seen · 433 lines · 19 tokens per session scan A 6cabee8a21c2

Subscribe to this mod's changes

advanced-code-review-verify is a command published in the GitHub repository axiomantic/spellbook (10 stars, last pushed yesterday), licensed MIT. It adds 19 tokens to every session and 3,926 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.