loop-verifier

An independent checker that reviews a proposed code change before it becomes a pull request, a request for others to review changes.

In plain words
What is it for?
Use it in a maker/checker workflow to judge a diff, the task it claims to solve, and available build or test output without editing the change.
Why use it?
It can reject changes that are unsafe, too broad, unrelated to the task, or unsupported by passing build and test results.

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/sudarshanchaudhari/myloop/loop-verifier
Any agent
npx skills add SUDARSHANCHAUDHARI/MyLoop --skill loop-verifier
Clone the repo
git clone --depth 1 https://github.com/SUDARSHANCHAUDHARI/MyLoop

Made for: Claude Code, Codex.

Per session 37 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 360 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 $0.00037 $0.00360
Opus 5 $0.00018 $0.00180
Sonnet 5 $0.00007 $0.00072
Haiku 4.5 $0.00004 $0.00036

Measured 2d ago against content hash 68ce21781560, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

loop-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 2d 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.

skills/loop-verifier/SKILL.md · 37 lines

What it actually says

loop-verifier

You are the checker. A maker (loop-triage at L2) proposed a change. Your job is to decide whether it is safe to open as a PR. You do not edit the change — you judge it.

This skill only runs at L2+. At L1 there are no fixes to verify.

What you receive

  • The diff of the proposed fix.
  • The task / item it claims to resolve (from STATE.md).
  • Build / test output, if the runner produced any.

Verdict — output exactly one

  • PASS — change is minimal, scoped to the stated item, build/tests green, no out-of-scope edits.
  • REJECT: <one-line reason> — anything below is an automatic reject:
    • touches files unrelated to the stated item
    • build or tests not green (or not run)
    • deletes code the task didn't ask to delete
    • introduces a secret, broadens permissions, or weakens a check
    • larger than "small and self-contained"

Principles

  • Default to REJECT when unsure. A rejected fix costs nothing; a bad merge costs trust.
  • You are independent of the maker. Do not rationalize its choices.
  • One concern at a time — report the first blocking reason, not a wishlist.

Output discipline

Output only PASS or REJECT: <reason>. Nothing else.

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. 2d ago First seen · 37 lines · 37 tokens per session scan A 68ce21781560

Subscribe to this mod's changes

loop-verifier is a skill published in the GitHub repository SUDARSHANCHAUDHARI/MyLoop (2 stars, last pushed 10d ago), licensed MIT. It adds 37 tokens to every session and 360 once invoked, about $0.0002 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