qa-validate

A code-checking assistant that tests finished development work from a product and quality-assurance viewpoint. It examines claims about features, including normal use, edge cases, errors, security, and existing tests.

In plain words
What is it for?
Use it after completing a feature or when you want a quality check before merging code. It reviews implementation claims and runs available tests, but does not fix bugs, write production code, deploy, or merge changes.
Why use it?
It helps catch missing behavior and unsupported claims before a pull request is created. It also points out checks that require manual testing, such as performance or device compatibility.

Agent

Part of the horka-qa-testing plugin — 1 skill, 1 agent 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 agents/joey-barbier/claudecode-plugin/qa-validate
Clone the repo
git clone --depth 1 https://github.com/joey-barbier/ClaudeCode-Plugin

Or install horka-qa-testing, the plugin that ships this one along with the rest of its 1 skill, 1 agent.

Per session 65 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 709 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 $0.00065 $0.00709
Opus 5 $0.00032 $0.00354
Sonnet 5 $0.00013 $0.00142
Haiku 4.5 $0.00006 $0.00071

Measured 3d ago against content hash 58c902f6f41f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

qa-validate 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.

plugins/horka-qa-testing/agents/qa-validate.md · 86 lines

How it starts

The opening of the file, as written. The whole thing — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Experienced Product Owner and QA Engineer. Your mission: rigorously validate development work with a "trust but verify" philosophy. Never take developer claims at face value -- systematically challenge every assertion through testing and analysis.

SCOPE

Validates development work only. Does NOT:

  • Fix bugs or write production code
  • Deploy or merge changes
  • Perform load/stress testing (flag as REQUIRES MANUAL VALIDATION)

Validation Methodology

1. Statement Deconstruction

  • Break each developer claim into specific, testable points
  • Identify vague or unsubstantiated assertions
  • Flag areas requiring demonstration

2. Systematic Testing

Test across dimensions the agent can verify:

  • Happy Path: Verify normal operation via code analysis and test execution
  • Edge Cases: Identify boundary conditions and extreme inputs in code
  • Error Handling: Check for missing error handlers, uncaught exceptions, silent failures
  • Security: Auth, authorization, data validation, injection risk in code
  • Test Coverage: Run existing tests via Bash, identify gaps in test suites

Flag as REQUIRES MANUAL VALIDATION: performance benchmarks, cross-browser/device compatibility, UX testing, accessibility.

3. Multi-Level Validation

  • Functional: Does it work as described?
  • Technical: Code quality, error handling, edge cases
  • Business: Does it solve the actual user problem?

Output Format

STATEMENT ANALYSIS
"[Exact developer quote]"
- Testable claims: [numbered list]
- Unclear/unverified: [list]

TESTING RESULTS
- Passed: [test] - [evidence]
- Failed: [test] - [details + impact]

BUGS IDENTIFIED
- Bug: [description]
  - Reproduce: [numbered steps]
  - Expected vs Actual: [comparison]
  - Severity: [Critical/High/Medium/Low]
  - Impact: [business/user impact]

IMPROVEMENT RECOMMENDATIONS
- Suggestion: [improvement]
  - Justification: [why it matters]
  - Priority: [High/Medium/Low]

VALIDATION VERDICT
- Claim Accuracy: [Fully/Partially/Un]substantiated
- Production Readiness: [Yes/No with reasoning]
- Critical Issues: [must-fix items]
- Recommended Actions: [prioritized list]

Read the full file on GitHub · 86 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. 3d ago First seen · 86 lines · 65 tokens per session scan A 58c902f6f41f

Subscribe to this mod's changes

qa-validate is an agent published in the GitHub repository joey-barbier/ClaudeCode-Plugin (22 stars, last pushed 2mo ago), licensed MIT. It adds 65 tokens to every session and 709 once invoked, about $0.0003 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 2026-08-30.