Home / arcasilesgroup / ai-engineering · skills/ai-pr-loop-fix/SKILL.md · GitHub

ai-pr-loop-fix skillA

ai-pr-loop-fix is agent-read markdown (skill) from arcasilesgroup/ai-engineering: Post-implementation CI verification loop: pushes the current branch, polls CI via gh, diagnoses failures, applies obvious fixes (lint, import, type errors), and loops until green or 5 iterations cap. Never auto-merges. Trigger for "fix CI", "get to green", "CI loop", "verify CI". Not for diagnosing non-CI failures — use /ai-debug. Not for adding test coverage — use /ai-verify. Not for complex logic fixes — report and stop..

Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.

What the file says

# ai-pr-loop-fix — push, poll, fix obvious, report the rest

## What it produces

A CI-green branch, or a clear report of what failed and why the fix is non-trivial.

## Steps

1. **Push the current branch.** `git push`. If push fails (upstream not set), set it and
   push again. If the branch is `main` or `master`, refuse: this skill fixes feature
   branches, not trunks.

2. **Poll CI.** `gh run list --branch <branch> --limit 1 --json databaseId,status,conclusion`.
   If `status` is `completed` and `conclusion` is `success`, skip to step 7. If
   `status` is `in_progress` or `queued`, wait 30s and poll again. Max 3 polls before
   declaring CI stuck (report and stop).

3. **Read the failure.** `gh run view <run-id> --log-failed` to get the actual logs.
   Scan for the first actionable error. The rest are usually cascades.

4. **Diagnose.** Classify the failure:
   - **Obvious**: lint error, missing import, unused variable, type mismatch, formatting.
   - **Logic**: wrong test expectation, broken behaviour, subtle type issue.
   - **Infrastructure**: flaky test, network timeout, dependency resolution.
…

Read the whole file at its exact version.

How to install

Latest version
mdr add arcasilesgroup/ai-engineering/ai-pr-loop-fix@git:20260923.5ad45c5
Exact content
mdr add arcasilesgroup/ai-engineering/ai-pr-loop-fix@sha256:793a4196c860ffb5

Pin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_d6dsvjk52qgvoz4n.svg)](https://markdownregistry.com/a/art_d6dsvjk52qgvoz4n)

0 badge views in 30 days

Versions

versioncommittedcommitsizeaudit
git:20260923.5ad45c5 latest2026-09-23 5ad45c5 3,681 BA view · diff
git:20260923.767162b2026-09-23 767162b 3,627 BA view

Audit of the latest version

A  17 of 17 checks passed. Deterministic, no model, same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (3681 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

arcasilesgroup/ai-engineering · 58 stars · license Apache-2.0 · pushed 2026-09-24 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_d6dsvjk52qgvoz4n
GET https://markdownregistry.com/api/v1/resolve?ref=arcasilesgroup/ai-engineering/ai-pr-loop-fix
GET https://markdownregistry.com/api/v1/blob/793a4196c860ffb5b6f99c5a86f548864464389bce60566890d6cb23415c5a89

Your agent does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

More from arcasilesgroup/ai-engineering

AGENTS.md agents
arcasilesgroup/ai-engineering · AGENTS.md
git:20260923.5be86d2 · audit A · 58 stars
ai-agents-md skill
arcasilesgroup/ai-engineering · skills/ai-agents-md/SKILL.md · Creates, audits and maintains the AGENTS.md a repository owes its coding agents, following the agents.md convention…
git:20260922.83b942f · audit A · 58 stars
ai-architect skill
arcasilesgroup/ai-engineering · skills/ai-architect/SKILL.md · Research and advise on the best way to approach a software project, including architecture, tech stack, implementation…
git:20260910.d021330 · audit A · 58 stars
ai-audit-code skill
arcasilesgroup/ai-engineering · skills/ai-audit-code/SKILL.md · Forces the laziest solution that actually works, simplest, shortest, most minimal. Channels a senior dev who has seen…
git:20260923.767162b · audit A · 58 stars
ai-audit-design skill
arcasilesgroup/ai-engineering · skills/ai-audit-design/SKILL.md · Use when a web interface needs its visual defects found and fixed with measurements rather than opinions — misaligned…
git:20260923.5be86d2 · audit B · 58 stars
ai-audit-docs skill
arcasilesgroup/ai-engineering · skills/ai-audit-docs/SKILL.md · Audits documentation files against actual repo state. Detects dead references (links to nonexistent files/URLs), false…
git:20260923.5ad45c5 · audit A · 58 stars
ai-brainstorm skill
arcasilesgroup/ai-engineering · skills/ai-brainstorm/SKILL.md · Use when the user brings a fuzzy idea that must be pinned down before research or planning — "handshake", "get on the…
git:20260923.5be86d2 · audit A · 58 stars
ai-code-review skill
arcasilesgroup/ai-engineering · skills/ai-code-review/SKILL.md · Comprehensive code review skill. Two modes: diff review for over-engineering findings (ranked, compact), and debt…
git:20260923.5ad45c5 · audit A · 58 stars
ai-codegraph skill
arcasilesgroup/ai-engineering · skills/ai-codegraph/SKILL.md · Semantic code intelligence via MCP: call chains, blast radius, cross-file symbol relationships. Uses CodeGraph…
git:20260923.5ad45c5 · audit A · 58 stars
ai-debug skill
arcasilesgroup/ai-engineering · skills/ai-debug/SKILL.md · Finds the root cause of broken behaviour and names it at file:line, then writes the check that fails for that reason…
git:20260914.793714f · audit A · 58 stars
ai-design skill
arcasilesgroup/ai-engineering · skills/ai-design/SKILL.md · Entry point for the ai-engineering design skills. Works out what a design or frontend request actually needs, elects…
git:20260923.5be86d2 · audit A · 58 stars
ai-explore skill
arcasilesgroup/ai-engineering · skills/ai-explore/SKILL.md · Answers questions about this repository by reading it, anchored to file:line, and tours an unfamiliar area for somebody…
git:20260922.83b942f · audit A · 58 stars

Every file in arcasilesgroup/ai-engineering

Browse by kind, by grade A, or by owner.