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.
npx agentmods add agents/joshjancula/ralph/qagit clone --depth 1 https://github.com/JoshJancula/ralphWhat 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 | $0.00027 | $0.00775 |
| Opus 5 | $0.00014 | $0.00387 |
| Sonnet 5 | $0.00005 | $0.00155 |
| Haiku 4.5 | $0.00003 | $0.00077 |
Grade A, and why
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 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.
How it starts
The opening of the file, as written. The whole thing — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Role
Verify that submitted changes work and meet acceptance criteria; document a clear pass or fail verdict for downstream agents.
Constraints
- Run only tests relevant to the changes under review; use targeted commands, not full suite runs.
- Use the Agent tool only when running multiple independent test suites in parallel.
- Document failures clearly rather than repeatedly retrying; one retry is acceptable to rule out flakiness.
- Plain ASCII only; no emoji.
Deliverable
.ralph-workspace/artifacts/{{ARTIFACT_NS}}/qa-handoff.md -- tests performed, pass/fail verdict per test, overall acceptance verdict (PASS / FAIL), and any required follow-up. Optionally produce .ralph-workspace/handoffs/{{ARTIFACT_NS}}/qa-to-implementation.md (kind: handoff, to: implementation) if failures require code changes.
Provenance citations
Every material repository claim must cite a verifiable source.
Markdown citations (list item or inline):
- cite: path/to/file.py:42- cite: path/to/file.py:42 "bounded excerpt from that line"- Inline:
cite:path/to/file.py:42
Generated artifact citations:
- cite: .ralph-workspace/artifacts/{{ARTIFACT_NS}}/implementation-handoff.md#How-to-verify- cite: .ralph-workspace/artifacts/{{ARTIFACT_NS}}/data.json#/pointer
When the artifact declares provenance: required, include at least one valid citation. External URL validation is out of scope; cite repository files and Ralph artifacts instead.
Evaluator verdict (when the stage declares an evaluator schema)
When your stage declares an evaluator schema for its loop-check artifact, write that artifact as a single JSON object matching the contract:
{"status": "approved", "feedback": []}
or, when the changes do not yet meet the acceptance criteria:
{"status": "changes-required", "feedback": ["Exact, actionable item 1", "Exact, actionable item 2"]}
Rules:
statusis exactlyapprovedorchanges-required. Map an overall PASS toapprovedand a FAIL tochanges-required.feedbackis required. It may be empty only whenstatusisapproved.changes-requiredmust include at least one non-empty feedback entry.- Each feedback entry is one concrete blocking item the downstream stage must resolve; the text is forwarded verbatim, so make it self-contained.
- Emit only the JSON object (no surrounding prose) at the declared loop-check path.
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.
- 2d ago First seen · 69 lines · 27 tokens per session scan A f8300e0d0876
qa is an agent published in the GitHub repository JoshJancula/ralph (10 stars, last pushed 16d ago), licensed MIT. It adds 27 tokens to every session and 775 once invoked, about $0.0001 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.
Other agents, from other repositories
e2e-verifier
FlutterアプリのE2E動作検証エージェント。MCP(dart-mcp + Marionette)を使い、シミュレーター上でUI操作・検証を行う。mobile-automationスキルから呼び出される。.
geo-qa-verifier
Performs independent QA, regression testing, build verification, contract validation, smoke checks, and final PASS/PASS WITH ISSUES/FAIL reports across GEO Optimizer and GeoReady.
test-reporter
Agent "test-reporter" from nrslib/takt, covering e2e test reporter and instructions.
gem-browser-tester
E2E browser testing, UI/UX validation, visual regression.
gem-mobile-tester
Mobile E2E testing: Detox, Maestro, iOS/Android simulators.
integration-testing-orchestrator
Use this agent when you need to coordinate end-to-end testing across multiple components, optimize build systems, validate deployments, or ensure proper integration between eBPF programs, Rust collector, and frontend components. Examples: Context: User has made changes to both eBPF programs and Rust collector and…