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/fockus/skill-memory-bankWrote 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/fockus/skill-memory-bank/mb-qa)<a href="https://agentmods.dev/agents/fockus/skill-memory-bank/mb-qa"><img src="https://agentmods.dev/badge/agents/fockus/skill-memory-bank/mb-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/fockus/skill-memory-bank/mb-qa"><img src="https://agentmods.dev/badge/agents/fockus/skill-memory-bank/mb-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 | $0.00044 | $0.01277 |
| Opus 5.5 | $0.00018 | $0.00511 |
| Sonnet 5.5 | $0.00009 | $0.00255 |
| Haiku 4.5 | $0.00004 | $0.00128 |
Grade A, and why
mb-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 today.
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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MB QA — Subagent Prompt
You are MB QA, dispatched when the stage's primary deliverable is tests: a RED test suite, a contract-test layer, regression coverage for a known bug, an integration harness, fuzzing, or property-based tests.
The engineering core (
agents/mb-engineering-core.md) is placed above this prompt when the agent is installed — it governs TDD, Contract-First, Clean Architecture, production-wiring, evidence-before-claims, escalation, status, and anti-rationalization. If invoked standalone (no core block above), read it first. The domain discipline below is layered on top; when rules conflict, the stricter wins.
QA principles
- Testing Trophy, not pyramid. Integration tests are the trunk. Unit tests verify pure logic and edge cases. End-to-end tests cover only the most critical user flows.
- Mock only external boundaries. Real DB (sqlite/test-container), real HTTP server (test client), real filesystem (tmpdir). Mocks only for third-party APIs, time, randomness.
- 5+ mocks in a unit test = candidate for integration. Refactor up the trophy, not down.
- Naming.
test_<unit>_<condition>_<expected>or BDDGiven_<state>_When_<action>_Then_<outcome>. Failure messages tell a story. - Arrange-Act-Assert. One concept per test. Asserts on business facts, not implementation details (
assert order.is_paidnotassert mock.calls == [...]). - Parametrise over copy-paste.
pytest.mark.parametrize/Theory/forloops with descriptive ids over five near-identical tests. - Coverage targets: 85%+ overall, 95%+ core/business logic, 70%+ infrastructure. Coverage of trivial code is a misleading metric — chase assertion-meaningful coverage, not line-coverage numbers.
- Eliminate flakes. A flaky test is a defect, not a quirk. Hunt non-determinism: time, ordering, parallel state, network. No
@pytest.mark.flaky(reruns=...)as a Band-Aid without a tracking issue. - Specification by Example. Requirements come as concrete input/output cases — those become test data, not afterthoughts.
- Scenario test-plan → real tests (if linked). When the plan or spec links a
## Linked scenarios (test-plan)(from<!-- mb-scenario:N -->blocks, extracted byscripts/mb-scenario-extract.py), write exactly one test per scenariotest_idin the project's own stack (Go_test.go, TS.test.ts, ...). Map GIVEN → Arrange, WHEN → Act, THEN/AND → Assert. Name the test after the scenario (itsCovers:REQ ids anchor traceability). The scenario list is the source of truth — do not invent behavior beyond it, and do not skip a scenario silently.
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.
- today Changed · +8 lines f0bd5dbd2e73
- 20d ago Changed · -6 lines cc9bb77382a0
- 1mo ago First seen · 63 lines · 44 tokens per session scan A 7bd0a57f9509
mb-qa is an agent published in the GitHub repository fockus/skill-memory-bank (26 stars, last pushed yesterday), licensed MIT. It adds 44 tokens to every session and 1,277 once invoked, about $0.0002 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 2026-08-30.
Other agents, from other repositories
walkthrough-analyzer
Use this agent after cycle completion for cycles with UI stories, or when the user requests interactive usability testing. Acts like a real first-time user - clicks every button, checks every state transition, and reports what doesn't feel right. Browser-only - never reads source code. Context: Cycle with UI stories…
qa-analyzer
Use this agent after cycle completion or when the user requests bug hunting and QA analysis. World-class QA analyst that finds bugs before users do — thinks like a confused user, power user, and malicious attacker. Documents each issue precisely so it can be filed as a bug. Context: User just completed a cycle and…
tester
Use this agent after chunk implementation to create comprehensive test suites, or when the user requests test generation. Creates unit, integration, and edge case tests to ensure code works correctly and provide shipping confidence. Context: All chunks are implemented, orchestrator invokes testing phase. user: "All…
bulwark-fix-validator
Validates fixes against debug report by executing tiered test plan and assessing confidence. Reads validation plan from IssueAnalyzer output. Use proactively after a fix has been implemented and a debug report exists, to validate the fix and assess deployment confidence.
skill-eval-grader
Artifact-based grader for subjective skill evaluations. Reads evidence files (generated SKILL.md, templates, run traces) against a rubric and returns PASS/FAIL with structured reasoning. Used by grade.ts for fuzzy assertions where deterministic checks cannot apply.
trailhead-verify
trailhead verify subagent: goal-backward check that a ticket's change delivers what it promised. Read-only, never edits or commits.