qa-loop

An automated audit-and-fix loop for code changes. It runs several audits in parallel, checks findings for false positives, fixes high-priority issues, and verifies the result with tests and linting.

In plain words
What is it for?
Use it to review the difference between your branch and the main branch, automatically fix prioritized findings, run verification checks, or perform an audit-only or pull-request comment mode.
Why use it?
It helps catch real problems in a change while filtering out misleading findings and repeating checks until the stop conditions are met.

Agent for Claude Code

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/christopherlouet/claude-base/qa-loop
Clone the repo
git clone --depth 1 https://github.com/christopherlouet/claude-base

Made for: Claude Code.

Per session 72 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,770 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 $0.00072 $0.02770
Opus 5 $0.00036 $0.01385
Sonnet 5 $0.00014 $0.00554
Haiku 4.5 $0.00007 $0.00277

Measured 2d ago against content hash b737709018aa, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

qa-loop 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.

.claude/agents/qa-loop.md · 256 lines

How it starts

The opening of the file, as written. The whole thing — 256 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent QA-LOOP

Autonomous AUDIT (parallel) → VALIDATE → FIX → VERIFY → CHECK loop with stop criteria. Adopts the Anthropic 2026 pattern (official code-review plugin): parallelization, false positive validation, high-signal filter, auto-scope.

Global workflow

┌────────────────────────────────────────────────────────────────────────┐
│                         QA-LOOP (v2)                                    │
│                                                                         │
│   ┌──────────────┐   ┌──────────┐   ┌─────────┐   ┌──────────┐         │
│   │   AUDIT      │──→│ VALIDATE │──→│  FIX    │──→│  VERIFY  │         │
│   │ 4 sub-agents │   │ 1 per    │   │ P0 then │   │ tests    │         │
│   │ in parallel  │   │ finding  │   │ P1      │   │ lint     │         │
│   └──────────────┘   └──────────┘   └─────────┘   └──────────┘         │
│         ↑                                                │               │
│         │              ┌──────────┐                      │               │
│         └──────────────│  CHECK   │←─────────────────────┘               │
│                        │ stop     │                                      │
│                        │ criteria │                                      │
│                        └──────────┘                                      │
│                              │                                           │
│                  score >= target AND 0 P0/P1 ?                           │
│                              │                                           │
│                         YES: STOP                                        │
│                         NO: LOOP (back to AUDIT)                         │
└────────────────────────────────────────────────────────────────────────┘

Parameters

Parameter Default Description
Target score 90/100 Minimum score to stop the loop
Max iterations 5 Maximum number of audit-fix cycles
Domains all security, perf, a11y, claudemd (4 sub-agents)
Fix severity P0+P1 Only fix validated high-signal issues
Scope git diff main...HEAD Audit limited to files modified on the branch
--audit-only off Read-only mode: audit + report, no FIX
--comment off Post inline on the current PR via gh pr comment

Read the full file on GitHub · 256 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. 2d ago First seen · 256 lines · 72 tokens per session scan A b737709018aa

Subscribe to this mod's changes

qa-loop is an agent published in the GitHub repository christopherlouet/claude-base (5 stars, last pushed 2d ago), licensed MIT. It adds 72 tokens to every session and 2,770 once invoked, about $0.0004 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 agents, from other repositories

gsd-plan-checker

Verifies plans will achieve phase goal before execution. Goal-backward analysis of plan quality. Spawned by /gsd:plan-phase orchestrator.

travisjneuman/.claude · 36 tokens

go-expert

Go concurrency, error handling, stdlib patterns, Chi/Echo web frameworks specialist. Use when writing Go code, designing concurrent systems, or building Go web services. Trigger phrases: Go, Golang, goroutine, channel, Chi, Echo, stdlib, context, error handling, interface, module, go test.

travisjneuman/.claude · 69 tokens

architecture-analyst

Analyzes system architecture, identifies patterns/anti-patterns, and provides strategic recommendations. Use for architectural reviews, refactoring planning, or system design decisions.

travisjneuman/.claude · 36 tokens

blockchain-developer

Solidity, smart contracts, Web3, EVM, DeFi protocols, and blockchain development specialist. Use when writing smart contracts, auditing contract security, or building dApps. Trigger phrases: Solidity, smart contract, Web3, blockchain, EVM, DeFi, NFT, Ethereum, Hardhat, Foundry, ethers.js, wagmi.

travisjneuman/.claude · 76 tokens

microservices-architect

Expert microservices architect for distributed system design, service decomposition, and resilience patterns.

travisjneuman/.claude · 20 tokens

adapter_grok

Grok is an eagerly registered stock-TUI adapter. RimZ launches grok, installs passive global hooks in ${GROKHOME:-/.grok}/hooks/rimz.json, and enriches each session from its durable updates.jsonl, summary.json, signals.json, and optional events.jsonl files. ACP and provider-private billing APIs stay outside this…

rimio-ai/rimz · 0 tokens