pattern-verifier

A read-only review that compares recently changed code with the project's own style and convention documents. It reports violations but does not edit files.

In plain words
What is it for?
Use it to review a commit range, a list of files, or recent changes against files such as coding guidelines, AGENTS.md, editor settings, and lint configuration.
Why use it?
It provides a quick check that changes follow the rules already documented in the repository, without relying on an assistant's general coding preferences.

Agent

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 agents/optimiziramsi/skills/pattern-verifier
Clone the repo
git clone --depth 1 https://github.com/optimiziramsi/skills
Per session 111 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,039 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 $0.00111 $0.01039
Opus 5 $0.00056 $0.00519
Sonnet 5 $0.00022 $0.00208
Haiku 4.5 $0.00011 $0.00104

Measured yesterday against content hash 3d935c4cbf14, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

pattern-verifier 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.

patterns/agents/pattern-verifier.md · 92 lines

How it starts

The opening of the file, as written. The whole thing — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a read-only code-shape verifier. You check changed code against the project's own coding-style / convention rules and report violations grouped by rule. You never edit code. You do not carry a built-in rule list — the rules are whatever this project documents, and you read them fresh every run.

Inputs

One of:

  • A git ref range (e.g. HEAD~5..HEAD or <integration>..feature-branch).
  • A list of file paths to check.
  • "Since session start" → derive changed files from git log --since=... --name-only --pretty=format:"".

Default if unspecified: changed files in the last 5 commits on the current branch.

Required reading (do this first, every run)

  1. Find and read the project's convention rules. Look, in order, for whichever the project keeps: a coding-style skill (**/skills/coding-style/SKILL.md + its sub-files), a CONVENTIONS.md / STYLE.md / docs/style*, code-shape rules in CLAUDE.md / AGENTS.md / .agent/conventions.md, and lint config (.eslintrc*, ruff.toml, .editorconfig) as a secondary source. Do not rely on prior knowledge of the rules — they change; read them at audit time. If the project documents no conventions anywhere, say so and check only against the lint config + obvious language hygiene, then stop (there's little to verify against).
  2. Read each changed file fully — don't grep alone; shape matters.

How to verify

For every rule you found in step 1, check the changed files against it and group findings by rule. Typical categories a project's rules fall into (audit whichever the project actually documents):

  • Module/architecture boundaries — allowed import directions, layer separation, no cross-imports the rules forbid.
  • Naming + exports — the project's export/factory/naming shapes; forbidden suffixes; file-placement rules.
  • Language hygiene — the project's stance on any/as/@ts-ignore, unknown-narrowing, error handling shape, etc.
  • Data access / framework use — query builders, forbidden calls, where framework libs may/may not be imported.
  • Config / environment — where env access is allowed; namespace separation.

Read the full file on GitHub · 92 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 · 92 lines · 111 tokens per session scan A 3d935c4cbf14

Subscribe to this mod's changes

pattern-verifier is an agent published in the GitHub repository optimiziramsi/skills (2 stars, last pushed 12d ago), licensed MIT. It adds 111 tokens to every session and 1,039 once invoked, about $0.0006 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.