ef プラグイン

Stage 5 - adversarial audit of delivered work against its requirements, verification that claimed fixes are real, then the code-verified contract doc

Stage 2 - author the feature folder (RFC, PRD, INDEX) from closed grill decisions

Relentless one-question-at-a-time interview that closes every open decision before implementation starts - Stage 1 of a feature at product altitude, or the decision stage of a repo discovery

Stage 4 - extract checks with named proofs from the contract, then orchestrate the build in parallel slices

Stage 3 - discover a feature from THIS repo's perspective before implementing it - anchor check, scout fan-out, contradiction triage, then grill

Discover a brand-new UI before building it - 4-5 radically different variants in one shareable HTML file

Make the repo's agent rulebook verifiable - vendor the checker, then audit, bootstrap, or grow AGENTS.md from audit findings

Stage 3 pre-check - verify every code anchor in a feature's docs still resolves in this repo

Turn a feature into a small, product-level Jira issue set - draft first, push through the host's Jira tools only after explicit approval

Stage 5 - adversarially audit delivered work against its requirements assuming a green test suite proves nothing, then verify that claimed fixes are real, then emit the code-verified contract document other repositories build against. Use after implementing a feature in a repo, after remediating audit findings, or when handing a feature to another repo.

Stage 2 - author a feature's documentation folder from closed grill decisions: grill-decisions to RFC to PRD to INDEX, with mandatory fields gated by asking rather than invented, stable requirement identifiers, and an amendment protocol that keeps the documents alive through implementation. Use after a product-altitude grilling has closed the decisions, or when the user says write the RFC, write the PRD, author the feature docs.

Relentless one-question-at-a-time interview that converts an under-specified plan into closed decisions, each recorded with its evidence grade, preserved dissent, riskiest assumption and reopen condition. Runs at product altitude as Stage 1 of a feature, or at repo altitude as the decision step of a discovery. Use when the user says grill me, stress-test this plan, close the open decisions, or when a feature needs defining before anything is built. Do NOT use to write the RFC or PRD, or to implement.

Stage 4 - turn closed decisions and a written contract into checks with named proofs, then orchestrate the build in parallel slices and hand a whole feature to the audit. Use when discovery closed, the contract is written, or the user says implement this feature, build this ticket, extract a checklist, ef-implement. Do NOT use when nobody has decided what to build, or to design the work.

Discover a feature from the current repository's perspective, after it was defined upstream at product altitude: verify the docs' code anchors, fan out read-only scouts, triage contradictions, grill the survivors, then write the contract. Use when handed a feature-docs folder (grill-decisions/RFC/PRD) produced elsewhere, once per target repo, before implementing.

Discover a brand-new UI before implementing it: generate 4-5 radically different variants of one screen in a single shareable HTML file, switchable from a floating bar, then pick one and let it govern the requirements and the contract. Use when a feature introduces a screen, page or component that is not a variation of anything the repo already has. Do NOT use for forms, tables, list-detail, settings panes, modals or any well-known pattern, and do NOT use to prototype logic or state models.

Make a repository's agent rulebook (AGENTS.md / CLAUDE.md) verifiable: vendor a deterministic checker, audit every rule and citation against the code, bootstrap a lean rulebook from what the code cannot say, and grow it from audit findings. Use when setting a repo up for agents, when asked to audit or slim AGENTS.md or CLAUDE.md, before the first EF feature in a repo, or after an audit proposes rulebook lines. Do NOT use to generate a codebase overview or documentation tour.

Shared vocabulary for every ef-* stage: the A–D evidence grades and the five critical lenses (assumption-surfacing, pre-mortem, red-team, evidence-audit, second-order). Load when an ef-* skill tells you to grade a claim or apply a lens. Not a workflow on its own.

Turn any feature description into a small, product-level Jira issue set: settle project and language, size by semantic change, write a draft, and push through the host's own Jira tools only after explicit approval. Standalone and optional: no EF docs required, outside the five stages. Use when the user asks to create Jira issues, tickets or tasks for a feature, to align a feature to Jira, or to break a feature into Jira tasks. Do NOT use to write an RFC or PRD, to plan implementation, or to edit issues that already exist.

Edit|Write|MultiEdit