# Lean Quality Engine

> |

- **Type:** Skill
- **Install:** `agentstack add skill-r-irbe-proof-skills-lean-quality-engine`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [r-irbe](https://agentstack.voostack.com/s/r-irbe)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** Apache-2.0
- **Upstream author:** [r-irbe](https://github.com/r-irbe)
- **Source:** https://github.com/r-irbe/proof-skills/tree/main/skills/lean-quality-engine
- **Website:** https://github.com/r-irbe/proof-skills

## Install

```sh
agentstack add skill-r-irbe-proof-skills-lean-quality-engine
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# lean-quality-engine

> ⚠️ **MANDATORY** (hot-tier): the soundness and faithfulness gates in
> §Behavioural rules are hard gates — a failure HALTS the QA lifecycle.
> Skipping Persist = incomplete.

## Routing

- **USE FOR:** running the seven-dimension quality lifecycle (Q1 soundness, Q2 faithfulness, Q3 completeness, Q4 novelty, Q5 elegance, Q6 integration, Q7 documentation); computing the weighted `Q_score`; deciding go/no-go at a milestone.
- **DO NOT USE FOR:** running a single script (delegate to `@lean-enforcement`); reviewing one proof (delegate to `@lean-proof-review`); facilitating a council session (delegate to `@lean-review-council`).
- **TRIGGERS:** QA, quality gate, milestone check, project health, qa-engine, quality score, QA lifecycle.

## Behavioural rules (G-*)

- **G-1** (MUST): Q1 (soundness) and Q2 (faithfulness) MUST both pass before any other dimension is scored. [Trace: AC-01]
- **G-2** (MUST): A failed hard gate (Q1 or Q2) MUST halt the lifecycle and route to recovery, regardless of other dimension scores. [Trace: AC-02]
- **G-3** (MUST): Every dimension score MUST cite the underlying script or review record (no inferred scores). [Trace: AC-03]
- **G-4** (SHOULD): Soft gates (Q3, Q6, Q7) SHOULD report gaps but MAY proceed when the owner has accepted the risk in writing. [Trace: AC-04]
- **G-5** (MUST NOT): The engine MUST NOT re-implement checks owned by `@lean-enforcement`; it MUST call out to that skill. [Trace: AC-05]
- **G-6** (MUST): The weighted `Q_score` MUST be persisted to the repository's quality tracker with timestamp, commit, and per-dimension breakdown. [Trace: AC-06]
- **G-7** (SHOULD): When `Q_score` regresses ≥ 5 % vs prior milestone, the engine SHOULD escalate before proceeding. [Trace: AC-07]
- **G-8** (MUST): On any guard failure the skill MUST escalate per §Recovery & STOP. [Trace: AC-08]

## Workflow

1. **Discover** [discover] — read prior `Q_score`, list active milestone, locate latest enforcement / council / coverage artefacts.
2. **Run hard gates** [execute] — invoke `@lean-enforcement` for Q1 (axiom audit, sorry check) and Q2 (faithfulness via doc-requirements traceability).
3. **Halt-check** [validate] — if Q1 or Q2 failed, STOP and route to recovery; do NOT proceed to soft gates.
4. **Run soft gates** [execute] — coverage (Q3), novelty (Q4), elegance (Q5), integration (Q6), documentation (Q7) via the appropriate scripts and council snapshots.
5. **Score** [validate] — compute weighted `Q_score`; compare to prior milestone; flag regressions ≥ 5 %.
6. **Persist** [persist] *(MANDATORY)* — write the `Q_score` breakdown to the project tracker, update milestone status, tick the QA checklist. Skipping Persist = incomplete.

## Recovery & STOP

- Q1 or Q2 fails → immediate STOP; escalate to human; do NOT compute `Q_score`.
- Enforcement script crashes ×3 in one lifecycle → STOP, file a bug against `@lean-enforcement`.
- `Q_score` regression ≥ 10 % vs prior milestone → STOP, ask before persisting.
- Confidence  AI agents commonly: score soft gates before hard gates pass; persist a `Q_score`
> that includes a failed hard gate; re-implement an enforcement check inline
> instead of calling `@lean-enforcement`; skip the regression delta against prior
> milestone. Full registry: GUARDRAILS.md §Agent failure taxonomy.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [r-irbe](https://github.com/r-irbe)
- **Source:** [r-irbe/proof-skills](https://github.com/r-irbe/proof-skills)
- **License:** Apache-2.0
- **Homepage:** https://github.com/r-irbe/proof-skills

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-r-irbe-proof-skills-lean-quality-engine
- Seller: https://agentstack.voostack.com/s/r-irbe
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.