audit-code

A repeated code-review and repair workflow that continues until a review round finds no new issues. It requires an existing project setup and at least one completed code module.

In plain words
What is it for?
Use it to audit modules or a broader code theme, group work by module ownership, fix findings, and continue until the review becomes clean.
Why use it?
It reduces the chance that the first review leaves related problems unresolved. The workflow keeps reviewing and fixing the selected scope across multiple rounds.

Skill for Claude CodeCodex

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 skills/leifericf/agentic-sdk/audit-code
Any agent
npx skills add leifericf/agentic-sdk --skill audit-code
Clone the repo
git clone --depth 1 https://github.com/leifericf/agentic-sdk

Made for: Claude Code, Codex.

Per session 16 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 857 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 $0.00016 $0.00857
Opus 5 $0.00008 $0.00428
Sonnet 5 $0.00003 $0.00171
Haiku 4.5 $0.00002 $0.00086

Measured yesterday against content hash 74a585ead3ef, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

audit-code 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.

skills/audit-code/SKILL.md · 75 lines

How it starts

The opening of the file, as written. The whole thing — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.

audit-code

Role: the audit loop. Reviews and fixes existing code at any scope, running review rounds until a round finds nothing new, then offers the stack to land.

Prerequisites

~/.agentic-sdk/<project>/project.edn exists with :lanes, :dimensions-active, and :spine. At least one module has landed; there is code to audit.

Procedure

The per-round procedure is run-review-round; this skill is the loop around it. The topology and conflict rules are in skills/shared/references/worktree-model.md.

  1. Initialize. Start the audit change as the next commit on the current tip, never on the main bookmark: jj new, then jj describe -m "Refactor: Audit <scope>". Hold the scope (a module, a list of modules, or a theme) in context; the rest of the run cites it. A long audit may keep a minimal, gitignored resume checkpoint under ~/.agentic-sdk/<project>/runs/<slug>/ recording what is done (rounds completed, findings still open); never committed, never the hand-off medium.
  2. Shard the scope by module ownership: one shard is one module directory plus its tests. For a theme scope, grep the theme to a file list first, then group by module. Dispatch gather-module-context once per shard and reuse the briefs across all dimensions in the round.
  3. Rounds while findings warrant. Dispatch review-round-runner with the audit change id, the shards, and the round number. It runs run-review-round: fans out the active check-* dimensions over each shard, runs the spine triage task to dedupe and order findings, dispatches editor waves by level via apply-findings, and re-verifies. It returns one summary line; its escalations and out-of-scope items ride back in that line and you hold them. A full audit is not bound by the per-phase two-round cap implement-change uses: finding everything is the point. Keep going while a round still surfaces high- or medium-severity findings. A round that finds only low-severity or style findings is the last one: record those and stop rather than spend another full round. Cap the loop at a round budget stated up front (default four) to bound cost; say so if hit. Resume after a crash per the worktree-model resume law.
  4. Report. The audit change is already a linear stack on the tip; nothing to restack. Report from context: the rounds table (the one-line summaries), what was fixed by level, the escalations the rounds returned, and the merged changelog set (the editors' CHANGELOG: lines from each round summary). Offer the stack to land; advancing main is the maintainer's call, never autonomous.

Read the full file on GitHub · 75 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. yesterday First seen · 75 lines · 16 tokens per session scan A 74a585ead3ef

Subscribe to this mod's changes

audit-code is a skill published in the GitHub repository leifericf/agentic-sdk (5 stars, last pushed 13d ago), licensed MIT. It adds 16 tokens to every session and 857 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.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens