cli-audit-drift

A code review tool that compares the software with its intended behavior written in functional contracts, rules, and invariants. It looks for behavior that still runs but no longer matches those intentions.

In plain words
What is it for?
Use it before committing or while reviewing changes to compare implementation against CONTRACTS.md, check important rules, and find semantic drift.
Why use it?
It catches silent changes that ordinary compilation and tests may miss, especially when code produces valid results in the wrong way.

Skill for Claude CodeCodex

Part of the cli-code-skills plugin — 36 skills shipped together

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/destynova2/cli-code-skills/cli-audit-drift
Any agent
npx skills add Destynova2/cli-code-skills --skill cli-audit-drift
Clone the repo
git clone --depth 1 https://github.com/Destynova2/cli-code-skills

Made for: Claude Code, Codex.

Or install cli-code-skills, the plugin that ships this one along with the rest of its 36 skills.

Per session 135 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,304 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 $0.00135 $0.03304
Opus 5 $0.00068 $0.01652
Sonnet 5 $0.00027 $0.00661
Haiku 4.5 $0.00014 $0.00330

Measured 3d ago against content hash a81e356550a4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cli-audit-drift 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 3d 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.

cli-audit-drift/SKILL.md · 259 lines

How it starts

The opening of the file, as written. The whole thing — 259 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Optimization: This skill uses on-demand loading. Heavy content lives in references/ and is loaded only when needed.

Language rule: Skill instructions are written in English. When generating user-facing output (reports, files, documentation), detect the project's primary language (from README, comments, docs, commit messages) and produce the output in that language. If the project is bilingual, ask the user which language to use before proceeding.

Audit Drift — Semantic Conformity Scanner

Detect silent behavioral drift between what code was supposed to do (intention) and what it actually does (implementation). The most dangerous bugs are not crashes — they are functions that still run, still return values, but no longer honor their original contract.

"The New Caledonian crow does not merely endure its environment — it identifies the mismatch between what it intended to do and what actually happened. Not a noisy error. A result that does not match the intention."

Core Principle — Three biological mechanisms

Organism / mechanism What it does in nature What it does in this skill
New Caledonian crow Detects the gap between intention and result Compares CONTRACTS.md (intention) against code (behavior). Names the gap precisely. Asks if intentional or accidental
Cellular autophagy Continuous surveillance, isolates damaged components Scans all contracted functions, isolates non-conforming code, proposes minimal correction or contract update
Protein folding + chaperones A 1D amino-acid sequence folds into a 3D functional structure; chaperones (HSP60/70) detect misfolds and force a clean re-fold from scratch Contract (intent, 1D) "folds" into code (behavior, 3D); detected drift triggers a minimal re-derivation from the contract, never a patch on top of the drifted state. Removed contracts trigger explicit code deletion (ubiquitin tag)

Read the full file on GitHub · 259 lines

Files

What ships with it

4 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.

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. 3d ago First seen · 259 lines · 135 tokens per session scan A a81e356550a4

Subscribe to this mod's changes

cli-audit-drift is a skill published in the GitHub repository Destynova2/cli-code-skills (5 stars, last pushed 10d ago), licensed MIT. It adds 135 tokens to every session and 3,304 once invoked, about $0.0007 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.