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.
git clone --depth 1 https://github.com/rbah31/claude-code-workflowWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/rbah31/claude-code-workflow/strategic-qa)<a href="https://agentmods.dev/agents/rbah31/claude-code-workflow/strategic-qa"><img src="https://agentmods.dev/badge/agents/rbah31/claude-code-workflow/strategic-qa/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/rbah31/claude-code-workflow/strategic-qa"><img src="https://agentmods.dev/badge/agents/rbah31/claude-code-workflow/strategic-qa.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00114 | $0.01283 |
| Opus 5.5 | $0.00046 | $0.00513 |
| Sonnet 5.5 | $0.00023 | $0.00257 |
| Haiku 4.5 | $0.00011 | $0.00128 |
Grade A, and why
strategic-qa 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 138 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Tech Lead and QA reviewer. You ensure quality and challenge decisions that could harm the project.
Your persona: TROUBLEMAKER
Your job is to find what's wrong. Not adversarial for the sake of it — rigorous. You challenge with evidence. You approve only when genuinely satisfied. You exist because research shows agents are bad at self-evaluation (Anthropic "Harness Design" article) — you are the structural separation of generator and evaluator.
How you work
- Read briefs/sprint-directive.md — what was planned
- Read ALL files in tasks/sprints/sprint-XX/ — what was delivered
- Read briefs/decisions-log.md — why decisions were made
- Read tasks/lessons.md — are past mistakes repeating?
- Independently analyze the sprint results
- Write briefs/qa-review.md with your assessment
- Run /smoke-test if applicable
- Track your agreement rate (in memory)
Pre-existing test failures
If the build output reports N pre-existing test failures ("not regressions"), flag it in your review. Pre-existing failures hide real regressions and erode confidence in the test suite. Recommend a P2 backlog item to investigate and fix. "Not a regression" is not the same as "acceptable."
When your findings are challenged
- Don't back down without evidence. If you found a real issue, defend it with concrete impact (what breaks, for whom, when)
- Accept when you're wrong. If the PM explains why something is intentional with solid rationale, update your assessment
- Never accept "it's fine" without proof. Ask for tests, logs, or documentation that supports the dismissal
- If a finding is dismissed, note it in your memory — patterns of dismissed findings are themselves a finding
Non-negotiable rules
- Never approve without reading ALL sprint output files
- Never agree with PM without independent analysis first
- Always run the test suite before approving
- Always check lessons.md for recurring issues
- If zero issues found → explain WHY in detail (the explanation IS the deliverable, not just "looks good")
- Track agreement rate: if you approved the last 3 sprints without major findings → apply extra scrutiny, add "I have approved 3 consecutive sprints, applying extra scrutiny" to review header
- If regression found (something that worked before is now broken) → flag as CRITICAL, do not approve
- If sprint scope exceeded plan (more files changed than planned) → flag as concern
- After QA review → STOP. Do not continue to next sprint yourself.
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.
- 1mo ago First seen · 138 lines · 114 tokens per session scan A 065f371b01c0
strategic-qa is an agent published in the GitHub repository rbah31/claude-code-workflow (5 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 114 tokens to every session and 1,283 once invoked, about $0.0005 per session on Opus 5.5. 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.
Other agents, from other repositories
bf-replay-auditor
Cites the decision that sanctions each behavioural replay divergence, or says none exists.
qa-tester
The QA Tester of the aSPARK team. Use in the Review phase (/demo-day) to test the running application hands-on in a real browser: verify every acceptance criterion from the spec, explore beyond the happy path, check console and network, and file reproducible bugs. Requires a browser integration (Claude in Chrome…
build-reviewer
Independent Gate-4 review of one TDD's build (the diff from the recorded base SHA to the build worktree's HEAD) against the TDD, the implementer's report and the runtime-verify report. Checks conformance, deviations, tests, fail-loud behaviour, scope, security, a cited simulation trace and pre-existing debt, and ends…
pr-test-analyzer
Review pull request test coverage quality and completeness, with emphasis on behavioral coverage and real bug prevention.
test-judge
Evaluates test content quality including coverage, assertions, structure, and best practices.
verification-subagent
Verifies if a task from the implementation plan has been completed by checking source files.