[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79YM92V7F09XWBQYA7BTSAF":3,"skill-same:01M79YM92V7F09XWBQYA7BTSAF":60},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":29,"ownerHtmlUrl":30,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":31,"install":32,"resources":45},"01M79YM92V7F09XWBQYA7BTSAF","requirement-review",null,"Challenge a feature or story before SDD in one of two modes. Mode --feature checks functional gaps and acceptance criteria; mode --tech grounds contracts and flows in the repository. Use when reviewing requirements or invoking requirement-review.","tibursocampos\u002Fagent-dev-toolkit","tibursocampos","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F37385246?v=4","01M79YF4CE4TH665XJJ7XSHXXK","Raphael Campos","01M79YF4D4KDQK01HZ6835E4A0",6,"MIT","https:\u002F\u002Fgithub.com\u002Ftibursocampos\u002Fagent-dev-toolkit\u002Fblob\u002Fmaster\u002Fcore\u002Fskills\u002Frequirement-review\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: requirement-review\ndescription: Challenge a feature or story before SDD in one of two modes. Mode --feature checks functional gaps and acceptance criteria; mode --tech grounds contracts and flows in the repository. Use when reviewing requirements or invoking requirement-review.\n---\n\n## STOP - Read before ANY tool call\n\nRead `_shared\u002Fskill-boilerplate.md` first. It holds the guardrails read, the session gate rule, the user `sim` rule, the one-step rule, the gate check, and the user chat language rule. The steps below are skill-specific only.\n\nWrites (review report, Q&A history) are new artifacts: show the target path and summary, then wait for `sim` before any Write (`guardrails.md` section 2). Never write code.\n\n---\n\n# Skill: requirement-review\n\n## Trigger\n**Negative signal:** applies as defined in `_shared\u002Fskill-boilerplate.md`; the skill-specific signals are listed below.\n\nInvoke when the user asks for: `\u002Frequirement-review`, `requirement-review --feature`, `requirement-review --tech`, \"challenge this requirement\", \"review the acceptance criteria\", or \"check this story against the repository\".\n\nNot a trigger: a request to write a PRD (`sdd-spec`), to break steps into a checklist (`split-story-checklist`), or to change code (`developer`, `*-developer`).\n\n**Mode (mandatory, exactly one per run - no silent default):**\n\n| Mode | Explicit invoke examples | Playbook |\n|------|--------------------------|----------|\n| **--feature** | `--feature`, `feature`, `1` | `references\u002Ffeature.md` |\n| **--tech** | `--tech`, `tech`, `2` | `references\u002Ftech.md` |\n\nIf the invocation names neither mode, or both: **STOP** after gate check (-1) and ask once, in the **user chat language**, then wait. Load **no** playbook until the mode is set. Do not assume a mode.\n\nWhen `invocation_context=orchestrated` (a parent such as `refine-story` passes the mode), do not ask. A missing mode is `blocked`: return to the parent.\n\nQuestion component form (ids fixed in English; labels rendered in the user chat language):\n\n```text\nheader: requirement-review mode\nquestion: requirement-review mode?\nsingle select:\n- feature: functional review of the story or feature (invoke token --feature)\n- tech: repository-grounded technical review (invoke token --tech)\n```\n\nRender the question and the two labels in the user chat language. Keep the mode ids `feature` and `tech` in English. The invoke tokens `--feature` and `--tech` map to them.\n\n## Outcome\n\nOne review report per run, with:\n\n- Status token `READY` or `NEEDS_CLARIFICATION` (`_shared\u002Fsdd-artifacts\u002Freadiness-severity.md`).\n- Open questions with severity, decision owner, and one answerable question each.\n- Findings with stable ids (`refine-story\u002Freferences\u002Ffinding-format.md`).\n- Atomized acceptance criteria (feature mode) or grounding and follow-ups (tech mode).\n- Q&A entries appended to the Q&A history.\n\nFlow outcome per `_shared\u002Fskills-catalog\u002FCATALOG.md` (Flow terminal contract): `done` when READY and the handoff is named; `open_question` while any question waits for an answer; `blocked` when the source or the repository is missing.\n\n## Flow contract: reads, writes, where, when\n\n- **Reads:** `STORY.md` or `FEATURE.md`, `REFINE\u002Fqa-history.md` when present, or pasted text.\n- **Writes:** after `sim` only: `REFINE\u002Frequirement-review-feature.md` (or `-tech.md`) and `REFINE\u002Fqa-history.md`. Pasted source: chat only.\n- **Where:** `features\u002FNNN-slug\u002FUSnn\u002FREFINE\u002F`; without a story folder, `docs\u002Fbacklog\u002F\u003Cslug>-requirement-review-\u003Cmode>.md`.\n- **When to use:** challenge a feature or story before SDD, with `--feature` (acceptance criteria) or `--tech` (repository contracts).\n\nFull flow: [backlog refinement guide](..\u002F..\u002F..\u002Fdocs\u002Fguides\u002Fbacklog-refine.md).\n\n## Lazy-load\n\n| When | Path |\n|------|------|\n| Index (routing only) | `{{TOOLKIT_ROOT}}\u002Fskills\u002Frequirement-review\u002Freference.md` |\n| Mode playbook **feature** (only when mode = --feature) | `{{TOOLKIT_ROOT}}\u002Fskills\u002Frequirement-review\u002Freferences\u002Ffeature.md` |\n| Mode playbook **tech** (only when mode = --tech) | `{{TOOLKIT_ROOT}}\u002Fskills\u002Frequirement-review\u002Freferences\u002Ftech.md` |\n| API contract standards (tech mode, only when an API contract is affected) | `{{TOOLKIT_ROOT}}\u002Fskills\u002Fapi-standards\u002FSKILL.md`, then one matching file under its `references\u002F` |\n| Severity B \u002F I \u002F MINOR + READY gate | `{{TOOLKIT_ROOT}}\u002Fskills\u002F_shared\u002Fsdd-artifacts\u002Freadiness-severity.md` |\n| Clarify depth | `{{TOOLKIT_ROOT}}\u002Fskills\u002F_shared\u002Fbacklog-item-types\u002Fclarify-depth.md` |\n| Finding types and block shape | `{{TOOLKIT_ROOT}}\u002Fskills\u002Frefine-story\u002Freferences\u002Ffinding-format.md` |\n| Q&A history shape and flow rules | `{{TOOLKIT_ROOT}}\u002Fskills\u002Frefine-story\u002Freferences\u002Fqa-history.md` |\n| Storage paths and portable paths | `{{TOOLKIT_ROOT}}\u002Fskills\u002F_shared\u002Fsdd-artifacts\u002FSTORAGE.md` |\n| Invocation contexts (`direct` vs `orchestrated`) | `{{TOOLKIT_ROOT}}\u002Fskills\u002F_shared\u002Fsdd-artifacts\u002FINVOCATION-CONTEXTS.md` |\n| Selective retrieval (`SR-NO-FULL-DUMP`) | `{{TOOLKIT_ROOT}}\u002Fskills\u002F_shared\u002Fsdd-artifacts\u002FSELECTIVE-RETRIEVAL.md` |\n| Caveman Mode (if active) | `{{TOOLKIT_ROOT}}\u002Fskills\u002F_shared\u002Fcaveman\u002FCAVEMAN.md` - **Lite cap** |\n\n**Never by default:** do not preload `references\u002Ffeature.md` and `references\u002Ftech.md` together; load only the selected mode. Do not load `api-standards` unless the tech playbook finds an affected API contract. Do not dump `memory-bank\u002F`, a full PRD, or whole source files (`SR-NO-FULL-DUMP`).\n\n## Process\n\n### Step -1b - Caveman Mode (Lite cap)\n1. Read `{{SDD_ROOT}}\u002Fpreferences.json` (create `{ \"caveman_mode\": false, \"caveman_level\": \"full\" }` if missing).\n2. If `caveman_mode` is false: continue without compression.\n3. If true: load `{{TOOLKIT_ROOT}}\u002Fskills\u002F_shared\u002Fcaveman\u002FCAVEMAN.md`; apply the **Lite** cap and prefs `caveman_level`; show once: `[Caveman] Modo ativo (respostas compactas, level={effective}). Digite caveman off para desativar.`\n4. Honor `caveman on|off|status|lite|full|ultra` (and `stop caveman` \u002F `normal mode`) during the session.\n5. Auto-Clarity and never-compress gates, drafts, and paths per `CAVEMAN.md`.\n\n### 0. Workspace and invocation context\nConfirm the target repository only for `--tech`. Resolve `invocation_context` per `INVOCATION-CONTEXTS.md`: `direct` unless the parent handoff marks `orchestrated`.\n\n### 0.5 Mode\nResolve `--feature` or `--tech` from the invocation or from the answer to the mode prompt. Then read **only** that playbook.\n\n### 1. Locate source and prior history\n- **Source** (one of): `features\u002FNNN-slug\u002FUSnn\u002FSTORY.md` (or `TSnn`), `features\u002FNNN-slug\u002FFEATURE.md`, `REFINE\u002Frefine.md` beside the story, or pasted text. Pasted text has no file: keep the report in chat until the operator confirms a write path.\n- **Prior history first:** read `features\u002FNNN-slug\u002FUSnn\u002FREFINE\u002Fqa-history.md` (or the shortcut `docs\u002Fbacklog\u002F\u003Cslug>-qa-history.md`) before asking anything. Do not re-ask a question whose entry is `resolved` or `superseded`.\n- **Never erase a decision.** A changed decision is a new entry linked through `Supersedes` \u002F `Superseded by`, with the reason recorded in its Resolution field.\n\n### 2. Run the mode playbook\nFollow `references\u002Ffeature.md` or `references\u002Ftech.md`. It produces findings, candidate questions, and mode-specific sections. Shared rules below apply to both modes.\n\n### 3. Classify open questions\n- Severity **B**, **I**, or **MINOR** per `readiness-severity.md`. Depth per `clarify-depth.md`.\n- Decision owner: **Product** (value, scope, business rules, priority), **Engineering** (feasibility, contracts, data model, performance targets), or **UX** (flows, visible states, copy).\n- Each question has **one** answerable question. Prefer a closed choice with stable ids (for example `confirm`, `adjust`, `cancel`, or the domain ids listed in the question), labels rendered in the user chat language. Free text maps to `adjust` when it changes the proposal; otherwise the question stays open. Answer mapping: `{{TOOLKIT_ROOT}}\u002Fskills\u002F_shared\u002Fagents\u002FQUESTIONS.md`.\n- A product decision is never solved by assumption. Do not write a default into a criterion. Ask.\n- Ask every open question in one batch at the end, including MINOR. Silence does not close a question. \"I do not know\" leaves it open.\n\nQuestion entry (chat and report):\n\n```text\nQ-003 [I] owner: Product\nQuestion: \u003Cone answerable question>\nImpact: \u003Cportable path>#\u003Csection>; CA-002.a\n```\n\n### 4. Atomize acceptance criteria\nApplies to every criterion the source carries.\n\n- Keep the parent id (`CA-002`). When one criterion carries several independently observable outcomes, derive atoms `CA-002.a`, `CA-002.b`, and so on.\n- Never renumber an existing id. A new atom takes the next free suffix under its parent.\n- Each atom has Given \u002F When \u002F Then with an observable Then. Include the boundary atom (empty, minimum, maximum, duplicate, where the domain has them) and the failure atom (what the user sees and what is persisted).\n- An atom must be checkable by someone who did not write it. Reject \"works correctly\", \"as expected\", \"properly\", and \"funciona corretamente\" as Then.\n\n### 5. Report and Q&A\nOutput shape (report body; prose follows the user chat language, tokens stay English):\n\n```markdown\n# Requirement review - \u003Ctitle>\n\n| Field | Value |\n|-------|-------|\n| **Mode** | --feature \\| --tech |\n| **Invocation context** | direct \\| orchestrated |\n| **Source** | \u003Cportable path or \"(pasted text)\"> |\n| **Updated** | YYYY-MM-DD |\n\nStatus: READY | NEEDS_CLARIFICATION\n\n## Open questions\n\n| ID | Severity | Owner | Question |\n|----|----------|-------|----------|\n| Q-003 | I | Product | \u003Cone answerable question> |\n\n(Open items only. Resolved and superseded items live in the Q&A history.)\n\nAt every `open_question` stop and at close, you must show this full table in chat with every open question row (ID, Severity, Owner, Question), localized per `_shared\u002Fsdd-artifacts\u002FLIVE-STAGE-TABLE.md`. Never show only the changed row or summarize rows. A question the operator has resolved leaves the table and stays only in the Q&A history.\n\n## Findings\n\n[FND-001 | I | GAP] \u003Csection heading>\nEvidence: \u003Cportable path> | no-evidence\n\u003Cwhat is ambiguous, missing, contradictory, risky, or dependent>\nRecommendation: \u003Crecommendation, not a decision>\n\n## \u003Cmode section>\n\u003Cfeature: atomized criteria table; tech: grounding and follow-ups, see the playbook>\n\n## Next\n\u003Chandoff line from step 6>\n```\n\n- Finding ids, types, and block shape: `refine-story\u002Freferences\u002Ffinding-format.md`. Do not invent a second severity vocabulary.\n- Q&A entries: append new questions with `Status: open` **before** waiting for answers, using the `qa-history.md` shape. Set `Origin` to `requirement-review --feature` or `--tech`. Keep question ids stable; allocate the next free `Q-` and `D-` ids.\n- Persistence (after `sim`): `features\u002FNNN-slug\u002FUSnn\u002FREFINE\u002Frequirement-review-feature.md` or `requirement-review-tech.md`, with the Q&A history at `REFINE\u002Fqa-history.md`. Shortcut when no story folder exists: `docs\u002Fbacklog\u002F\u003Cslug>-requirement-review-\u003Cmode>.md` with `docs\u002Fbacklog\u002F\u003Cslug>-qa-history.md`. Source is a FEATURE.md with no story folder: report in chat and persist only to a story folder the operator names, or the shortcut. Do not invent a feature-root `REFINE\u002F` folder.\n- Portable paths only in the file, the handoff, and the history. No OS absolute paths.\n\n### 6. Status and handoff\nStatus follows `readiness-severity.md`: `READY` only with no unanswered B, I, or MINOR question. Recompute it on every refresh.\n\n| Status | Handoff |\n|--------|---------|\n| `READY`, mode `--feature`, no repository involved | `sdd-spec - \u003Cportable-story-path>` |\n| `READY`, mode `--feature`, story touches a repository | `requirement-review --tech - \u003Cportable-story-path>` next, then `sdd-spec` |\n| `READY`, mode `--tech` | `sdd-spec - \u003Cportable-story-path>`, with the listed technical follow-ups |\n| `NEEDS_CLARIFICATION` | Answer every open question, cite the Q&A history path, re-run the same mode. Do not claim the item is ready for `sdd-spec` |\n\nTechnical follow-ups are not questions. They never block READY; they travel to `sdd-spec` as a list.\n\n## Must not\n\n- Decide a product question by assumption, or write a default into an acceptance criterion\n- Claim READY while any unanswered B, I, or MINOR question remains\n- Write code, edit application files, or write a PRD or PLAN\n- Assert a repository fact without a search hit (`--tech`); record `not-found` instead of guessing\n- Copy the `api-standards` rules into the report; cite the reference file\n- Load both mode playbooks, or load the api-standards pack without an affected API contract\n- Erase or rewrite a prior answer, resolution, or impact; supersede it with a linked entry\n- Put more than one question in one entry\n- Call an external tracker, work-item API, or MCP work-item tool\n- Write OS absolute paths or InstallRoot embeds into the report, history, or handoff\n- Describe the design by reference to any external product, repository, or skill; state behavior in toolkit terms\n\n## Handoff examples\n\n```\nrequirement-review --feature - features\u002F004-export\u002FUS01\u002FSTORY.md\nrequirement-review --tech - features\u002F004-export\u002FTS02\u002FSTORY.md\n```\n\n```\n\u002Fsdd-spec - features\u002F004-export\u002FUS01\u002FSTORY.md\n```\n\nCite `REFINE\u002Fqa-history.md` in the typed handoff when present. Do not paste the history into the slash line.\n","## STOP - Read before ANY tool call\n\nRead `_shared\u002Fskill-boilerplate.md` first. It holds the guardrails read, the session gate rule, the user `sim` rule, the one-step rule, the gate check, and the user chat language rule. The steps below are skill-specific only.\n\nWrites (review report, Q&A history) are new artifacts: show the target path and summary, then wait for `sim` before any Write (`guardrails.md` section 2). Never write code.\n\n---\n\n# Skill: requirement-review\n\n## Trigger\n**Negative signal:** applies as defined in `_shared\u002Fskill-boilerplate.md`; the skill-specific signals are listed below.\n\nInvoke when the user asks for: `\u002Frequirement-review`, `requirement-review --feature`, `requirement-review --tech`, \"challenge this requirement\", \"review the acceptance criteria\", or \"check this story against the repository\".\n\nNot a trigger: a request to write a PRD (`sdd-spec`), to break steps into a checklist (`split-story-checklist`), or to change code (`developer`, `*-developer`).\n\n**Mode (mandatory, exactly one per run - no silent default):**\n\n| Mode | Explicit invoke examples | Playbook |\n|------|--------------------------|----------|\n| **--feature** | `--feature`, `feature`, `1` | `references\u002Ffeature.md` |\n| **--tech** | `--tech`, `tech`, `2` | `references\u002Ftech.md` |\n\nIf the invocation names neither mode, or both: **STOP** after gate check (-1) and ask once, in the **user chat language**, then wait. Load **no** playbook until the mode is set. Do not assume a mode.\n\nWhen `invocation_context=orchestrated` (a parent such as `refine-story` passes the mode), do not ask. A missing mode is `blocked`: return to the parent.\n\nQuestion component form (ids fixed in English; labels rendered in the user chat language):\n\n```text\nheader: requirement-review mode\nquestion: requirement-review mode?\nsingle select:\n- feature: functional review of the story or feature (invoke token --feature)\n- tech: repository-grounded technical review (invoke token --tech)\n```\n\nRender the question and the two labels in the user chat language. Keep the mode ids `feature` and `tech` in English. The invoke tokens `--feature` and `--tech` map to them.\n\n## Outcome\n\nOne review report per run, with:\n\n- Status token `READY` or `NEEDS_CLARIFICATION` (`_shared\u002Fsdd-artifacts\u002Freadiness-severity.md`).\n- Open questions with severity, decision owner, and one answerable question each.\n- Findings with stable ids (`refine-story\u002Freferences\u002Ffinding-format.md`).\n- Atomized acceptance criteria (feature mode) or grounding and follow-ups (tech mode).\n- Q&A entries appended to the Q&A history.\n\nFlow outcome per `_shared\u002Fskills-catalog\u002FCATALOG.md` (Flow terminal contract): `done` when READY and the handoff is named; `open_question` while any question waits for an answer; `blocked` when the source or the repository is missing.\n\n## Flow contract: reads, writes, where, when\n\n- **Reads:** `STORY.md` or `FEATURE.md`, `REFINE\u002Fqa-history.md` when present, or pasted text.\n- **Writes:** after `sim` only: `REFINE\u002Frequirement-review-feature.md` (or `-tech.md`) and `REFINE\u002Fqa-history.md`. Pasted source: chat only.\n- **Where:** `features\u002FNNN-slug\u002FUSnn\u002FREFINE\u002F`; without a story folder, `docs\u002Fbacklog\u002F\u003Cslug>-requirement-review-\u003Cmode>.md`.\n- **When to use:** challenge a feature or story before SDD, with `--feature` (acceptance criteria) or `--tech` (repository contracts).\n\nFull flow: [backlog refinement guide](..\u002F..\u002F..\u002Fdocs\u002Fguides\u002Fbacklog-refine.md).\n\n## Lazy-load\n\n| When | Path |\n|------|------|\n| Index (routing only) | `{{TOOLKIT_ROOT}}\u002Fskills\u002Frequirement-review\u002Freference.md` |\n| Mode playbook **feature** (only when mode = --feature) | `{{TOOLKIT_ROOT}}\u002Fskills\u002Frequirement-review\u002Freferences\u002Ffeature.md` |\n| Mode playbook **tech** (only when mode = --tech) | `{{TOOLKIT_ROOT}}\u002Fskills\u002Frequirement-review\u002Freferences\u002Ftech.md` |\n| API contract standards (tech mode, only when an API contract is affected) | `{{TOOLKIT_ROOT}}\u002Fskills\u002Fapi-standards\u002FSKILL.md`, then one matching file under its `references\u002F` |\n| Severity B \u002F I \u002F MINOR + READY gate | `{{TOOLKIT_ROOT}}\u002Fskills\u002F_shared\u002Fsdd-artifacts\u002Freadiness-severity.md` |\n| Clarify depth | `{{TOOLKIT_ROOT}}\u002Fskills\u002F_shared\u002Fbacklog-item-types\u002Fclarify-depth.md` |\n| Finding types and block shape | `{{TOOLKIT_ROOT}}\u002Fskills\u002Frefine-story\u002Freferences\u002Ffinding-format.md` |\n| Q&A history shape and flow rules | `{{TOOLKIT_ROOT}}\u002Fskills\u002Frefine-story\u002Freferences\u002Fqa-history.md` |\n| Storage paths and portable paths | `{{TOOLKIT_ROOT}}\u002Fskills\u002F_shared\u002Fsdd-artifacts\u002FSTORAGE.md` |\n| Invocation contexts (`direct` vs `orchestrated`) | `{{TOOLKIT_ROOT}}\u002Fskills\u002F_shared\u002Fsdd-artifacts\u002FINVOCATION-CONTEXTS.md` |\n| Selective retrieval (`SR-NO-FULL-DUMP`) | `{{TOOLKIT_ROOT}}\u002Fskills\u002F_shared\u002Fsdd-artifacts\u002FSELECTIVE-RETRIEVAL.md` |\n| Caveman Mode (if active) | `{{TOOLKIT_ROOT}}\u002Fskills\u002F_shared\u002Fcaveman\u002FCAVEMAN.md` - **Lite cap** |\n\n**Never by default:** do not preload `references\u002Ffeature.md` and `references\u002Ftech.md` together; load only the selected mode. Do not load `api-standards` unless the tech playbook finds an affected API contract. Do not dump `memory-bank\u002F`, a full PRD, or whole source files (`SR-NO-FULL-DUMP`).\n\n## Process\n\n### Step -1b - Caveman Mode (Lite cap)\n1. Read `{{SDD_ROOT}}\u002Fpreferences.json` (create `{ \"caveman_mode\": false, \"caveman_level\": \"full\" }` if missing).\n2. If `caveman_mode` is false: continue without compression.\n3. If true: load `{{TOOLKIT_ROOT}}\u002Fskills\u002F_shared\u002Fcaveman\u002FCAVEMAN.md`; apply the **Lite** cap and prefs `caveman_level`; show once: `[Caveman] Modo ativo (respostas compactas, level={effective}). Digite caveman off para desativar.`\n4. Honor `caveman on|off|status|lite|full|ultra` (and `stop caveman` \u002F `normal mode`) during the session.\n5. Auto-Clarity and never-compress gates, drafts, and paths per `CAVEMAN.md`.\n\n### 0. Workspace and invocation context\nConfirm the target repository only for `--tech`. Resolve `invocation_context` per `INVOCATION-CONTEXTS.md`: `direct` unless the parent handoff marks `orchestrated`.\n\n### 0.5 Mode\nResolve `--feature` or `--tech` from the invocation or from the answer to the mode prompt. Then read **only** that playbook.\n\n### 1. Locate source and prior history\n- **Source** (one of): `features\u002FNNN-slug\u002FUSnn\u002FSTORY.md` (or `TSnn`), `features\u002FNNN-slug\u002FFEATURE.md`, `REFINE\u002Frefine.md` beside the story, or pasted text. Pasted text has no file: keep the report in chat until the operator confirms a write path.\n- **Prior history first:** read `features\u002FNNN-slug\u002FUSnn\u002FREFINE\u002Fqa-history.md` (or the shortcut `docs\u002Fbacklog\u002F\u003Cslug>-qa-history.md`) before asking anything. Do not re-ask a question whose entry is `resolved` or `superseded`.\n- **Never erase a decision.** A changed decision is a new entry linked through `Supersedes` \u002F `Superseded by`, with the reason recorded in its Resolution field.\n\n### 2. Run the mode playbook\nFollow `references\u002Ffeature.md` or `references\u002Ftech.md`. It produces findings, candidate questions, and mode-specific sections. Shared rules below apply to both modes.\n\n### 3. Classify open questions\n- Severity **B**, **I**, or **MINOR** per `readiness-severity.md`. Depth per `clarify-depth.md`.\n- Decision owner: **Product** (value, scope, business rules, priority), **Engineering** (feasibility, contracts, data model, performance targets), or **UX** (flows, visible states, copy).\n- Each question has **one** answerable question. Prefer a closed choice with stable ids (for example `confirm`, `adjust`, `cancel`, or the domain ids listed in the question), labels rendered in the user chat language. Free text maps to `adjust` when it changes the proposal; otherwise the question stays open. Answer mapping: `{{TOOLKIT_ROOT}}\u002Fskills\u002F_shared\u002Fagents\u002FQUESTIONS.md`.\n- A product decision is never solved by assumption. Do not write a default into a criterion. Ask.\n- Ask every open question in one batch at the end, including MINOR. Silence does not close a question. \"I do not know\" leaves it open.\n\nQuestion entry (chat and report):\n\n```text\nQ-003 [I] owner: Product\nQuestion: \u003Cone answerable question>\nImpact: \u003Cportable path>#\u003Csection>; CA-002.a\n```\n\n### 4. Atomize acceptance criteria\nApplies to every criterion the source carries.\n\n- Keep the parent id (`CA-002`). When one criterion carries several independently observable outcomes, derive atoms `CA-002.a`, `CA-002.b`, and so on.\n- Never renumber an existing id. A new atom takes the next free suffix under its parent.\n- Each atom has Given \u002F When \u002F Then with an observable Then. Include the boundary atom (empty, minimum, maximum, duplicate, where the domain has them) and the failure atom (what the user sees and what is persisted).\n- An atom must be checkable by someone who did not write it. Reject \"works correctly\", \"as expected\", \"properly\", and \"funciona corretamente\" as Then.\n\n### 5. Report and Q&A\nOutput shape (report body; prose follows the user chat language, tokens stay English):\n\n```markdown\n# Requirement review - \u003Ctitle>\n\n| Field | Value |\n|-------|-------|\n| **Mode** | --feature \\| --tech |\n| **Invocation context** | direct \\| orchestrated |\n| **Source** | \u003Cportable path or \"(pasted text)\"> |\n| **Updated** | YYYY-MM-DD |\n\nStatus: READY | NEEDS_CLARIFICATION\n\n## Open questions\n\n| ID | Severity | Owner | Question |\n|----|----------|-------|----------|\n| Q-003 | I | Product | \u003Cone answerable question> |\n\n(Open items only. Resolved and superseded items live in the Q&A history.)\n\nAt every `open_question` stop and at close, you must show this full table in chat with every open question row (ID, Severity, Owner, Question), localized per `_shared\u002Fsdd-artifacts\u002FLIVE-STAGE-TABLE.md`. Never show only the changed row or summarize rows. A question the operator has resolved leaves the table and stays only in the Q&A history.\n\n## Findings\n\n[FND-001 | I | GAP] \u003Csection heading>\nEvidence: \u003Cportable path> | no-evidence\n\u003Cwhat is ambiguous, missing, contradictory, risky, or dependent>\nRecommendation: \u003Crecommendation, not a decision>\n\n## \u003Cmode section>\n\u003Cfeature: atomized criteria table; tech: grounding and follow-ups, see the playbook>\n\n## Next\n\u003Chandoff line from step 6>\n```\n\n- Finding ids, types, and block shape: `refine-story\u002Freferences\u002Ffinding-format.md`. Do not invent a second severity vocabulary.\n- Q&A entries: append new questions with `Status: open` **before** waiting for answers, using the `qa-history.md` shape. Set `Origin` to `requirement-review --feature` or `--tech`. Keep question ids stable; allocate the next free `Q-` and `D-` ids.\n- Persistence (after `sim`): `features\u002FNNN-slug\u002FUSnn\u002FREFINE\u002Frequirement-review-feature.md` or `requirement-review-tech.md`, with the Q&A history at `REFINE\u002Fqa-history.md`. Shortcut when no story folder exists: `docs\u002Fbacklog\u002F\u003Cslug>-requirement-review-\u003Cmode>.md` with `docs\u002Fbacklog\u002F\u003Cslug>-qa-history.md`. Source is a FEATURE.md with no story folder: report in chat and persist only to a story folder the operator names, or the shortcut. Do not invent a feature-root `REFINE\u002F` folder.\n- Portable paths only in the file, the handoff, and the history. No OS absolute paths.\n\n### 6. Status and handoff\nStatus follows `readiness-severity.md`: `READY` only with no unanswered B, I, or MINOR question. Recompute it on every refresh.\n\n| Status | Handoff |\n|--------|---------|\n| `READY`, mode `--feature`, no repository involved | `sdd-spec - \u003Cportable-story-path>` |\n| `READY`, mode `--feature`, story touches a repository | `requirement-review --tech - \u003Cportable-story-path>` next, then `sdd-spec` |\n| `READY`, mode `--tech` | `sdd-spec - \u003Cportable-story-path>`, with the listed technical follow-ups |\n| `NEEDS_CLARIFICATION` | Answer every open question, cite the Q&A history path, re-run the same mode. Do not claim the item is ready for `sdd-spec` |\n\nTechnical follow-ups are not questions. They never block READY; they travel to `sdd-spec` as a list.\n\n## Must not\n\n- Decide a product question by assumption, or write a default into an acceptance criterion\n- Claim READY while any unanswered B, I, or MINOR question remains\n- Write code, edit application files, or write a PRD or PLAN\n- Assert a repository fact without a search hit (`--tech`); record `not-found` instead of guessing\n- Copy the `api-standards` rules into the report; cite the reference file\n- Load both mode playbooks, or load the api-standards pack without an affected API contract\n- Erase or rewrite a prior answer, resolution, or impact; supersede it with a linked entry\n- Put more than one question in one entry\n- Call an external tracker, work-item API, or MCP work-item tool\n- Write OS absolute paths or InstallRoot embeds into the report, history, or handoff\n- Describe the design by reference to any external product, repository, or skill; state behavior in toolkit terms\n\n## Handoff examples\n\n```\nrequirement-review --feature - features\u002F004-export\u002FUS01\u002FSTORY.md\nrequirement-review --tech - features\u002F004-export\u002FTS02\u002FSTORY.md\n```\n\n```\n\u002Fsdd-spec - features\u002F004-export\u002FUS01\u002FSTORY.md\n```\n\nCite `REFINE\u002Fqa-history.md` in the typed handoff when present. Do not paste the history into the slash line.\n",{"name":5,"description":7},"master","d199760b9c98388be72a799dafccb9e981aa6d5f","core\u002Fskills\u002Frequirement-review",0,"https:\u002F\u002Fgithub.com\u002Ftibursocampos\u002Fagent-dev-toolkit","https:\u002F\u002Ftibursocampos.github.io\u002Fagent-dev-toolkit\u002F","https:\u002F\u002Fgithub.com\u002Ftibursocampos","2026-11-08 02:36:34",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add tibursocampos\u002Fagent-dev-toolkit --skill requirement-review",{"tool":38,"command":39},"bunx","bunx skills add tibursocampos\u002Fagent-dev-toolkit --skill requirement-review",{"tool":41,"command":42},"pnpm","pnpm dlx skills add tibursocampos\u002Fagent-dev-toolkit --skill requirement-review","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Ftibursocampos\u002Fagent-dev-toolkit\u002Fblob\u002Fmaster\u002Fcore\u002Fskills\u002Frequirement-review\u002FSKILL.md",true,[46,52,56],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"reference.md","REFERENCE","参考文档",944,"https:\u002F\u002Fgithub.com\u002Ftibursocampos\u002Fagent-dev-toolkit\u002Fblob\u002Fmaster\u002Fcore\u002Fskills\u002Frequirement-review\u002Freference.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Ffeature.md",4314,"https:\u002F\u002Fgithub.com\u002Ftibursocampos\u002Fagent-dev-toolkit\u002Fblob\u002Fmaster\u002Fcore\u002Fskills\u002Frequirement-review\u002Freferences\u002Ffeature.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Ftech.md",4604,"https:\u002F\u002Fgithub.com\u002Ftibursocampos\u002Fagent-dev-toolkit\u002Fblob\u002Fmaster\u002Fcore\u002Fskills\u002Frequirement-review\u002Freferences\u002Ftech.md",[61,68,75,82,89,96],{"skillNo":62,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":65,"categories":66,"keywords":67,"publishedAt":19},"01M79YNFMWYYP2QKV43VKJ90K0","user-own-skill","Operator-owned skill that is not part of the toolkit.","https:\u002F\u002Fgithub.com\u002Ftibursocampos\u002Fagent-dev-toolkit\u002Fblob\u002Fmaster\u002Fscripts\u002Fvalidation\u002Ffixtures\u002Fseed-user-root\u002Fskills\u002Fuser-own-skill\u002FSKILL.md",[],[],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":19},"01M79YNCCEJ2J9MF2Q1WYXE265","vue-developer","Implement or fix small-to-medium Vue 3 features without full SDD (Composition API, Pinia, configured project test runner; Vitest is a fallback for new projects). Use for isolated Vue work or when invoking \u002Fvue-developer.","https:\u002F\u002Fgithub.com\u002Ftibursocampos\u002Fagent-dev-toolkit\u002Fblob\u002Fmaster\u002Fcore\u002Fskills\u002Fvue-developer\u002FSKILL.md",[],[],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":19},"01M79YN85A2VGFD1P359F77YZ8","test-coverage","Run .NET Coverlet coverage, report Sonar-aligned metrics, and evaluate against a threshold (default 80%). Use for coverage reports or when invoking \u002Ftest-coverage.","https:\u002F\u002Fgithub.com\u002Ftibursocampos\u002Fagent-dev-toolkit\u002Fblob\u002Fmaster\u002Fcore\u002Fskills\u002Ftest-coverage\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":19},"01M79YN42PG910XDKK6AKBJB1H","split-story-checklist","Break refined backlog steps into a dependency-aware task checklist (backend, frontend, tests). Prefer features\u002F... story folders. Use when breaking down tasks or invoking \u002Fsplit-story-checklist.","https:\u002F\u002Fgithub.com\u002Ftibursocampos\u002Fagent-dev-toolkit\u002Fblob\u002Fmaster\u002Fcore\u002Fskills\u002Fsplit-story-checklist\u002FSKILL.md",[],[],{"skillNo":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":19},"01M79YMZKTTY1NSH39AYN6YSYZ","sdd-spec","Create a PRD for a new feature or change (agent PRD .md; language per LANGUAGE.md + preferences under features\u002F). Feeds sdd-plan. Use when creating a spec or invoking \u002Fsdd-spec.","https:\u002F\u002Fgithub.com\u002Ftibursocampos\u002Fagent-dev-toolkit\u002Fblob\u002Fmaster\u002Fcore\u002Fskills\u002Fsdd-spec\u002FSKILL.md",[],[],{"skillNo":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"publishedAt":19},"01M79YMV0YHHGCG527JEGY4RN0","sdd-plan","Create a baby-step PLAN from an existing PRD (agent PLAN .md; language per LANGUAGE.md + preferences). Feeds sdd-develop. Use when creating a plan or invoking \u002Fsdd-plan.","https:\u002F\u002Fgithub.com\u002Ftibursocampos\u002Fagent-dev-toolkit\u002Fblob\u002Fmaster\u002Fcore\u002Fskills\u002Fsdd-plan\u002FSKILL.md",[],[]]