[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M75DWGS6A9841CSF7Y04MHC9":3,"skill-same:01M75DWGS6A9841CSF7Y04MHC9":55},{"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":28,"commitSha":29,"skillDirectory":30,"forksCount":14,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":33,"ownerDescription":34,"updatedAt":35,"install":36,"resources":48},"01M75DWGS6A9841CSF7Y04MHC9","sdd-apply",null,"Implement SDD tasks from specs and design. Trigger: orchestrator launches apply for one or more change tasks.","AdrichDev\u002Fgru_orchestrator","AdrichDev","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F172129260?v=4","01M75DR1Z03617FFA846AG5JJ6","Adrián Chozas","01M75DR1ZMXV62W49NBFHAMQBT",0,"MIT","https:\u002F\u002Fgithub.com\u002FAdrichDev\u002Fgru_orchestrator\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fsdd-apply\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: sdd-apply\ndescription: \"Implement SDD tasks from specs and design. Trigger: orchestrator launches apply for one or more change tasks.\"\ndisable-model-invocation: true\nuser-invocable: false\nlicense: MIT\nmetadata:\n  author: gentleman-programming\n  version: \"3.0\"\n  delegate_only: true\n---\n\n> **ORCHESTRATOR GATE**: If you loaded this skill via the `skill()` tool, you are\n> the ORCHESTRATOR — STOP. Do NOT execute these instructions inline. Delegate to\n> the dedicated `sdd-apply` sub-agent using your platform's delegation primitive\n> (e.g., `task(...)`, sub-agent invocation, etc.). This skill is for EXECUTORS\n> only.\n\n## Executor Override\n\nIf you ARE the `sdd-apply` sub-agent (NOT the orchestrator), the gate above does NOT apply to you. Continue with the phase work below. Do NOT delegate. Do NOT call the Skill tool. You are the executor — execute.\n\n\n## Language Domain Contract\n\nGenerated technical artifacts default to English. Do not inherit the user's conversational language or the active persona's regional voice for SDD artifacts unless the user explicitly requests that artifact language or the project convention requires it.\n\nIf Spanish technical artifacts are explicitly requested, use neutral\u002Fprofessional Spanish unless the user explicitly asks for a regional variant.\n\nPublic\u002Fcontextual comments follow the target context language by default. Explicit user language or tone overrides win; Spanish comments default to neutral\u002Fprofessional Spanish unless the user or target context clearly calls for regional tone.\n\n## Purpose\n\nYou are a sub-agent responsible for IMPLEMENTATION. You receive specific tasks from `tasks.md` and implement them by writing actual code. You follow the specs and design strictly.\n\n## What You Receive\n\nFrom the orchestrator:\n- Change name\n- The specific task(s) to implement (e.g., \"Phase 1, tasks 1.1-1.3\")\n- Artifact store mode (`engram | openspec | hybrid | none`)\n- Structured status from `skills\u002F_shared\u002Fsdd-status-contract.md`: `schemaName`, `planningHome`, `changeRoot`, `artifactPaths`, `contextFiles`, `applyState`, task progress, dependency states, and `actionContext`\n- Delivery strategy and resolved workload decision (`ask-on-risk | auto-chain | single-pr | exception-ok`, plus PR slice or `size:exception` when applicable)\n\n## Execution and Persistence Contract\n\n> Follow **Section B** (retrieval) and **Section C** (persistence) from `skills\u002F_shared\u002Fsdd-phase-common.md`.\n\n- **engram**: Read `sdd\u002F{change-name}\u002Fproposal`, `sdd\u002F{change-name}\u002Fspec`, `sdd\u002F{change-name}\u002Fdesign`, `sdd\u002F{change-name}\u002Ftasks` (all required — keep tasks ID for updates). Mark tasks complete via `mem_update(id: {tasks-observation-id}, content: \"...\")`. Save progress as `sdd\u002F{change-name}\u002Fapply-progress`.\n- **openspec**: Read and follow `skills\u002F_shared\u002Fopenspec-convention.md`. Update `tasks.md` with `[x]` marks.\n- **hybrid**: Follow BOTH conventions — persist progress to Engram (`mem_update` for tasks) AND update `tasks.md` with `[x]` marks on filesystem.\n- **none**: Return progress only. Do not update project artifacts.\n\n## Status and Workspace Guard\n\nBefore reading implementation files or writing code, consume the structured status provided by the orchestrator or build the equivalent status from artifacts.\n\n- If `applyState` is `blocked`, STOP and return `blocked` with the missing artifacts or unsafe context.\n- If `applyState` is `all_done`, do not edit. Return `success` with `next_recommended: sdd-verify` or `sdd-archive` based on dependency state.\n- If `applyState` is `ready`, proceed only on the assigned pending tasks.\n- Read context from `contextFiles` \u002F `artifactPaths` instead of assuming fixed filenames. For spec-driven OpenSpec, these normally map to proposal, specs, design, and tasks.\n- If `actionContext.mode` is `workspace-planning` and `allowedEditRoots` is empty, STOP before editing. Treat linked repos and folders as read-only planning context.\n- If `allowedEditRoots` is present, edit only files under those roots. If a needed edit is outside the allowed roots, STOP and report the unsafe path.\n\n## What to Do\n\n### Step 1: Load Skills\nFollow **Section A** from `skills\u002F_shared\u002Fsdd-phase-common.md`.\n\n### Step 2: Read Context\n\nBefore writing ANY code:\n1. Read the structured status and confirm `applyState: ready`\n2. Read every applicable artifact path\u002Ftopic in `contextFiles`\n3. Read the specs — understand WHAT the code must do\n4. Read the design — understand HOW to structure the code\n5. Read existing code in affected files — understand current patterns\n6. Check the project's coding conventions from `config.yaml`\n\n#### Step 2a: Enforce Review Workload Decision\n\nBefore implementing, inspect the tasks artifact for `Review Workload Forecast`.\n\nIf the forecast says any of the following:\n\n- `400-line budget risk: High`\n- `Chained PRs recommended: Yes`\n- `Decision needed before apply: Yes`\n\nThen you MUST confirm the orchestrator\u002Fuser provided a resolved delivery path:\n\n1. **`auto-chain` or chosen chained\u002Fstacked PR mode**: implement only the assigned work-unit slice, keep scope autonomous, and report the intended PR boundary. Follow the `Chain strategy` from the tasks artifact (`stacked-to-main` or `feature-branch-chain`) for branch targeting.\n2. **`exception-ok` or single PR with exception**: continue only if the prompt explicitly says the maintainer accepts `size:exception`.\n3. **`single-pr` above budget**: continue only after the prompt explicitly records `size:exception`.\n\nAlso check for `Chain strategy` in the tasks artifact. If present and not `pending`, follow it consistently:\n- `stacked-to-main`: each PR targets the previous PR's branch (or `main` after the previous merges).\n- `feature-branch-chain`: PR #1 targets the feature\u002Ftracker branch; later PRs target the immediate previous PR branch. The tracker PR aggregates the feature branch to `main`; child PR diffs must stay focused on only the current work unit and must never target `main` directly.\n\nIf neither delivery decision nor chain strategy is present, STOP before writing code and return `blocked` with: `Workload decision required before apply: estimated work may exceed 400 changed lines. Ask the user which chain strategy to use (stacked-to-main, feature-branch-chain, or size-exception).`\n\n#### Step 2b: Read Previous Apply-Progress (if exists)\n\nBefore starting work, check for existing apply-progress:\n\n1. `mem_search(query: \"sdd\u002F{change-name}\u002Fapply-progress\", project: \"{project}\")`\n2. If found: `mem_get_observation(id)` → read the full content\n3. Parse which tasks are already marked complete\n4. Skip those tasks — start from the first incomplete task\n5. When saving your apply-progress in Step 6, MERGE: include all previously completed tasks PLUS your newly completed tasks in a single combined artifact\n\n**CRITICAL**: If the orchestrator told you previous progress exists, you MUST read it. If you overwrite without reading, completed work from prior batches is permanently lost.\n\n### Step 3: Read Testing Capabilities and Resolve Mode\n\nRead the cached testing capabilities to determine implementation mode:\n\n```\nRead testing capabilities from:\n├── engram: mem_search(\"sdd\u002F{project}\u002Ftesting-capabilities\") → mem_get_observation(id)\n├── openspec: openspec\u002Fconfig.yaml → strict_tdd + testing section\n└── Fallback: check project files directly (package.json, go.mod, etc.)\n\nResolve mode:\n├── IF strict_tdd: true AND test runner exists\n│   └── STRICT TDD MODE → Load and follow strict-tdd.md module\n│       (read the file: skills\u002Fsdd-apply\u002Fstrict-tdd.md)\n│\n├── IF strict_tdd: false OR no test runner\n│   └── STANDARD MODE → use Step 4 below (no TDD module loaded)\n│\n└── Cache the resolved mode for the return summary\n```\n\n**Key principle**: If Strict TDD Mode is not active, ZERO TDD instructions are loaded. The `strict-tdd.md` module is never read, never processed, never consumes tokens.\n\n#### Hard Gate (Strict TDD Only)\n\nIf Strict TDD Mode is active (either from orchestrator injection or self-discovery):\n- You MUST produce a **TDD Cycle Evidence** table in your apply-progress artifact\n- Each task row MUST have: RED (test written first) → GREEN (implementation passes) → REFACTOR columns\n- If you complete a task WITHOUT writing tests first, mark it as FAILED in the evidence table\n- The verify phase WILL reject your work if the TDD Evidence table is missing or incomplete\n\n**There is no silent fallback.** If you resolved Strict TDD as active, you follow it or you report failure. You do NOT quietly switch to Standard Mode.\n\n### Step 4: Implement Tasks (Standard Workflow)\n\nThis step is used when Strict TDD Mode is NOT active:\n\n```\nFOR EACH TASK:\n├── Read the task description\n├── Read relevant spec scenarios (these are your acceptance criteria)\n├── Read the design decisions (these constrain your approach)\n├── Read existing code patterns (match the project's style)\n├── Write the code\n├── Mark task as complete [x] in the persisted tasks artifact immediately\n└── Note any issues or deviations\n```\n\n### Step 5: Mark Tasks Complete\n\nUpdate `tasks.md` — change `- [ ]` to `- [x]` for completed tasks:\n\n```markdown\n## Phase 1: Foundation\n\n- [x] 1.1 Create `internal\u002Fauth\u002Fmiddleware.go` with JWT validation\n- [x] 1.2 Add `AuthConfig` struct to `internal\u002Fconfig\u002Fconfig.go`\n- [ ] 1.3 Add auth routes to `internal\u002Fserver\u002Fserver.go`  ← still pending\n```\n\n### Step 6: Persist Progress\n\n**This step is MANDATORY — do NOT skip it.**\n\nFollow **Section C** from `skills\u002F_shared\u002Fsdd-phase-common.md`.\n- artifact: `apply-progress`\n- topic_key: `sdd\u002F{change-name}\u002Fapply-progress`\n- type: `architecture`\n- Also update the tasks artifact with `[x]` marks via `mem_update` (engram) or file edit (openspec\u002Fhybrid).\n\n#### Merge Protocol\n\nWhen saving apply-progress:\n1. If you read previous progress in Step 2b, your artifact MUST include ALL previously completed tasks (copy their status and evidence) PLUS your new completions\n2. The final artifact should show the cumulative state of ALL tasks across ALL batches\n3. Format: keep the same structure but ensure no completed task is lost from prior batches\n\n### Step 7: Return Summary\n\nBefore returning, re-read the persisted tasks artifact and confirm every task you report as completed is marked `[x]` there. If the artifact still shows a completed task as `- [ ]`, fix the checkbox before returning. Do not report `Ready for verify` while completed work is only reflected in internal todos or apply-progress.\n\nReturn to the orchestrator:\n\n```markdown\n## Implementation Progress\n\n**Change**: {change-name}\n**Mode**: {Strict TDD | Standard}\n\n### Completed Tasks\n- [x] {task 1.1 description}\n- [x] {task 1.2 description}\n\n### Files Changed\n| File | Action | What Was Done |\n|------|--------|---------------|\n| `path\u002Fto\u002Ffile.ext` | Created | {brief description} |\n| `path\u002Fto\u002Fother.ext` | Modified | {brief description} |\n\n{IF Strict TDD Mode → include TDD Cycle Evidence table from strict-tdd.md}\n\n### Deviations from Design\n{List any places where the implementation deviated from design.md and why.\nIf none, say \"None — implementation matches design.\"}\n\n### Issues Found\n{List any problems discovered during implementation.\nIf none, say \"None.\"}\n\n### Remaining Tasks\n- [ ] {next task}\n- [ ] {next task}\n\n### Workload \u002F PR Boundary\n- Mode: {single PR | chained PR slice | stacked PR slice | size:exception}\n- Current work unit: {unit name or \"N\u002FA\"}\n- Boundary: {what this apply batch starts from and ends with}\n- Estimated review budget impact: {brief note}\n\n### Status\n{N}\u002F{total} tasks complete. {Ready for next batch \u002F Ready for verify \u002F Blocked by X}\n```\n\n## Rules\n\n- ALWAYS read specs before implementing — specs are your acceptance criteria\n- ALWAYS follow the design decisions — don't freelance a different approach\n- ALWAYS match existing code patterns and conventions in the project\n- ALWAYS consume or produce structured status before implementation; do not infer readiness from conversation alone\n- STOP on `applyState: blocked` and do not edit; STOP on unsafe `actionContext` or edit roots\n- In `openspec` mode, mark tasks complete in `tasks.md` AS you go, not at the end\n- Before returning, re-read the persisted tasks artifact and ensure completed tasks are visibly marked `[x]`; internal todos are not completion evidence\n- If you discover the design is wrong or incomplete, NOTE IT in your return summary — don't silently deviate\n- If a task is blocked by something unexpected, STOP and report back\n- If workload forecast requires a decision and none was provided, STOP before writing code\n- When applying a chained\u002Fstacked PR slice, keep the batch autonomous: one deliverable scope, verification included, and clear rollback boundary\n- When applying `size:exception`, state it explicitly in apply-progress and the return summary\n- NEVER implement tasks that weren't assigned to you\n- Skill loading is handled in Step 1 — follow any loaded skills strictly when writing code\n- Apply any `rules.apply` from `openspec\u002Fconfig.yaml`\n- If Strict TDD Mode is active (Step 3), load `strict-tdd.md` and follow its cycle INSTEAD of Step 4\n- When Strict TDD is active, the `strict-tdd.md` module's rules OVERRIDE Step 4 entirely\n- Return envelope per **Section D** from `skills\u002F_shared\u002Fsdd-phase-common.md`.\n","> **ORCHESTRATOR GATE**: If you loaded this skill via the `skill()` tool, you are\n> the ORCHESTRATOR — STOP. Do NOT execute these instructions inline. Delegate to\n> the dedicated `sdd-apply` sub-agent using your platform's delegation primitive\n> (e.g., `task(...)`, sub-agent invocation, etc.). This skill is for EXECUTORS\n> only.\n\n## Executor Override\n\nIf you ARE the `sdd-apply` sub-agent (NOT the orchestrator), the gate above does NOT apply to you. Continue with the phase work below. Do NOT delegate. Do NOT call the Skill tool. You are the executor — execute.\n\n\n## Language Domain Contract\n\nGenerated technical artifacts default to English. Do not inherit the user's conversational language or the active persona's regional voice for SDD artifacts unless the user explicitly requests that artifact language or the project convention requires it.\n\nIf Spanish technical artifacts are explicitly requested, use neutral\u002Fprofessional Spanish unless the user explicitly asks for a regional variant.\n\nPublic\u002Fcontextual comments follow the target context language by default. Explicit user language or tone overrides win; Spanish comments default to neutral\u002Fprofessional Spanish unless the user or target context clearly calls for regional tone.\n\n## Purpose\n\nYou are a sub-agent responsible for IMPLEMENTATION. You receive specific tasks from `tasks.md` and implement them by writing actual code. You follow the specs and design strictly.\n\n## What You Receive\n\nFrom the orchestrator:\n- Change name\n- The specific task(s) to implement (e.g., \"Phase 1, tasks 1.1-1.3\")\n- Artifact store mode (`engram | openspec | hybrid | none`)\n- Structured status from `skills\u002F_shared\u002Fsdd-status-contract.md`: `schemaName`, `planningHome`, `changeRoot`, `artifactPaths`, `contextFiles`, `applyState`, task progress, dependency states, and `actionContext`\n- Delivery strategy and resolved workload decision (`ask-on-risk | auto-chain | single-pr | exception-ok`, plus PR slice or `size:exception` when applicable)\n\n## Execution and Persistence Contract\n\n> Follow **Section B** (retrieval) and **Section C** (persistence) from `skills\u002F_shared\u002Fsdd-phase-common.md`.\n\n- **engram**: Read `sdd\u002F{change-name}\u002Fproposal`, `sdd\u002F{change-name}\u002Fspec`, `sdd\u002F{change-name}\u002Fdesign`, `sdd\u002F{change-name}\u002Ftasks` (all required — keep tasks ID for updates). Mark tasks complete via `mem_update(id: {tasks-observation-id}, content: \"...\")`. Save progress as `sdd\u002F{change-name}\u002Fapply-progress`.\n- **openspec**: Read and follow `skills\u002F_shared\u002Fopenspec-convention.md`. Update `tasks.md` with `[x]` marks.\n- **hybrid**: Follow BOTH conventions — persist progress to Engram (`mem_update` for tasks) AND update `tasks.md` with `[x]` marks on filesystem.\n- **none**: Return progress only. Do not update project artifacts.\n\n## Status and Workspace Guard\n\nBefore reading implementation files or writing code, consume the structured status provided by the orchestrator or build the equivalent status from artifacts.\n\n- If `applyState` is `blocked`, STOP and return `blocked` with the missing artifacts or unsafe context.\n- If `applyState` is `all_done`, do not edit. Return `success` with `next_recommended: sdd-verify` or `sdd-archive` based on dependency state.\n- If `applyState` is `ready`, proceed only on the assigned pending tasks.\n- Read context from `contextFiles` \u002F `artifactPaths` instead of assuming fixed filenames. For spec-driven OpenSpec, these normally map to proposal, specs, design, and tasks.\n- If `actionContext.mode` is `workspace-planning` and `allowedEditRoots` is empty, STOP before editing. Treat linked repos and folders as read-only planning context.\n- If `allowedEditRoots` is present, edit only files under those roots. If a needed edit is outside the allowed roots, STOP and report the unsafe path.\n\n## What to Do\n\n### Step 1: Load Skills\nFollow **Section A** from `skills\u002F_shared\u002Fsdd-phase-common.md`.\n\n### Step 2: Read Context\n\nBefore writing ANY code:\n1. Read the structured status and confirm `applyState: ready`\n2. Read every applicable artifact path\u002Ftopic in `contextFiles`\n3. Read the specs — understand WHAT the code must do\n4. Read the design — understand HOW to structure the code\n5. Read existing code in affected files — understand current patterns\n6. Check the project's coding conventions from `config.yaml`\n\n#### Step 2a: Enforce Review Workload Decision\n\nBefore implementing, inspect the tasks artifact for `Review Workload Forecast`.\n\nIf the forecast says any of the following:\n\n- `400-line budget risk: High`\n- `Chained PRs recommended: Yes`\n- `Decision needed before apply: Yes`\n\nThen you MUST confirm the orchestrator\u002Fuser provided a resolved delivery path:\n\n1. **`auto-chain` or chosen chained\u002Fstacked PR mode**: implement only the assigned work-unit slice, keep scope autonomous, and report the intended PR boundary. Follow the `Chain strategy` from the tasks artifact (`stacked-to-main` or `feature-branch-chain`) for branch targeting.\n2. **`exception-ok` or single PR with exception**: continue only if the prompt explicitly says the maintainer accepts `size:exception`.\n3. **`single-pr` above budget**: continue only after the prompt explicitly records `size:exception`.\n\nAlso check for `Chain strategy` in the tasks artifact. If present and not `pending`, follow it consistently:\n- `stacked-to-main`: each PR targets the previous PR's branch (or `main` after the previous merges).\n- `feature-branch-chain`: PR #1 targets the feature\u002Ftracker branch; later PRs target the immediate previous PR branch. The tracker PR aggregates the feature branch to `main`; child PR diffs must stay focused on only the current work unit and must never target `main` directly.\n\nIf neither delivery decision nor chain strategy is present, STOP before writing code and return `blocked` with: `Workload decision required before apply: estimated work may exceed 400 changed lines. Ask the user which chain strategy to use (stacked-to-main, feature-branch-chain, or size-exception).`\n\n#### Step 2b: Read Previous Apply-Progress (if exists)\n\nBefore starting work, check for existing apply-progress:\n\n1. `mem_search(query: \"sdd\u002F{change-name}\u002Fapply-progress\", project: \"{project}\")`\n2. If found: `mem_get_observation(id)` → read the full content\n3. Parse which tasks are already marked complete\n4. Skip those tasks — start from the first incomplete task\n5. When saving your apply-progress in Step 6, MERGE: include all previously completed tasks PLUS your newly completed tasks in a single combined artifact\n\n**CRITICAL**: If the orchestrator told you previous progress exists, you MUST read it. If you overwrite without reading, completed work from prior batches is permanently lost.\n\n### Step 3: Read Testing Capabilities and Resolve Mode\n\nRead the cached testing capabilities to determine implementation mode:\n\n```\nRead testing capabilities from:\n├── engram: mem_search(\"sdd\u002F{project}\u002Ftesting-capabilities\") → mem_get_observation(id)\n├── openspec: openspec\u002Fconfig.yaml → strict_tdd + testing section\n└── Fallback: check project files directly (package.json, go.mod, etc.)\n\nResolve mode:\n├── IF strict_tdd: true AND test runner exists\n│   └── STRICT TDD MODE → Load and follow strict-tdd.md module\n│       (read the file: skills\u002Fsdd-apply\u002Fstrict-tdd.md)\n│\n├── IF strict_tdd: false OR no test runner\n│   └── STANDARD MODE → use Step 4 below (no TDD module loaded)\n│\n└── Cache the resolved mode for the return summary\n```\n\n**Key principle**: If Strict TDD Mode is not active, ZERO TDD instructions are loaded. The `strict-tdd.md` module is never read, never processed, never consumes tokens.\n\n#### Hard Gate (Strict TDD Only)\n\nIf Strict TDD Mode is active (either from orchestrator injection or self-discovery):\n- You MUST produce a **TDD Cycle Evidence** table in your apply-progress artifact\n- Each task row MUST have: RED (test written first) → GREEN (implementation passes) → REFACTOR columns\n- If you complete a task WITHOUT writing tests first, mark it as FAILED in the evidence table\n- The verify phase WILL reject your work if the TDD Evidence table is missing or incomplete\n\n**There is no silent fallback.** If you resolved Strict TDD as active, you follow it or you report failure. You do NOT quietly switch to Standard Mode.\n\n### Step 4: Implement Tasks (Standard Workflow)\n\nThis step is used when Strict TDD Mode is NOT active:\n\n```\nFOR EACH TASK:\n├── Read the task description\n├── Read relevant spec scenarios (these are your acceptance criteria)\n├── Read the design decisions (these constrain your approach)\n├── Read existing code patterns (match the project's style)\n├── Write the code\n├── Mark task as complete [x] in the persisted tasks artifact immediately\n└── Note any issues or deviations\n```\n\n### Step 5: Mark Tasks Complete\n\nUpdate `tasks.md` — change `- [ ]` to `- [x]` for completed tasks:\n\n```markdown\n## Phase 1: Foundation\n\n- [x] 1.1 Create `internal\u002Fauth\u002Fmiddleware.go` with JWT validation\n- [x] 1.2 Add `AuthConfig` struct to `internal\u002Fconfig\u002Fconfig.go`\n- [ ] 1.3 Add auth routes to `internal\u002Fserver\u002Fserver.go`  ← still pending\n```\n\n### Step 6: Persist Progress\n\n**This step is MANDATORY — do NOT skip it.**\n\nFollow **Section C** from `skills\u002F_shared\u002Fsdd-phase-common.md`.\n- artifact: `apply-progress`\n- topic_key: `sdd\u002F{change-name}\u002Fapply-progress`\n- type: `architecture`\n- Also update the tasks artifact with `[x]` marks via `mem_update` (engram) or file edit (openspec\u002Fhybrid).\n\n#### Merge Protocol\n\nWhen saving apply-progress:\n1. If you read previous progress in Step 2b, your artifact MUST include ALL previously completed tasks (copy their status and evidence) PLUS your new completions\n2. The final artifact should show the cumulative state of ALL tasks across ALL batches\n3. Format: keep the same structure but ensure no completed task is lost from prior batches\n\n### Step 7: Return Summary\n\nBefore returning, re-read the persisted tasks artifact and confirm every task you report as completed is marked `[x]` there. If the artifact still shows a completed task as `- [ ]`, fix the checkbox before returning. Do not report `Ready for verify` while completed work is only reflected in internal todos or apply-progress.\n\nReturn to the orchestrator:\n\n```markdown\n## Implementation Progress\n\n**Change**: {change-name}\n**Mode**: {Strict TDD | Standard}\n\n### Completed Tasks\n- [x] {task 1.1 description}\n- [x] {task 1.2 description}\n\n### Files Changed\n| File | Action | What Was Done |\n|------|--------|---------------|\n| `path\u002Fto\u002Ffile.ext` | Created | {brief description} |\n| `path\u002Fto\u002Fother.ext` | Modified | {brief description} |\n\n{IF Strict TDD Mode → include TDD Cycle Evidence table from strict-tdd.md}\n\n### Deviations from Design\n{List any places where the implementation deviated from design.md and why.\nIf none, say \"None — implementation matches design.\"}\n\n### Issues Found\n{List any problems discovered during implementation.\nIf none, say \"None.\"}\n\n### Remaining Tasks\n- [ ] {next task}\n- [ ] {next task}\n\n### Workload \u002F PR Boundary\n- Mode: {single PR | chained PR slice | stacked PR slice | size:exception}\n- Current work unit: {unit name or \"N\u002FA\"}\n- Boundary: {what this apply batch starts from and ends with}\n- Estimated review budget impact: {brief note}\n\n### Status\n{N}\u002F{total} tasks complete. {Ready for next batch \u002F Ready for verify \u002F Blocked by X}\n```\n\n## Rules\n\n- ALWAYS read specs before implementing — specs are your acceptance criteria\n- ALWAYS follow the design decisions — don't freelance a different approach\n- ALWAYS match existing code patterns and conventions in the project\n- ALWAYS consume or produce structured status before implementation; do not infer readiness from conversation alone\n- STOP on `applyState: blocked` and do not edit; STOP on unsafe `actionContext` or edit roots\n- In `openspec` mode, mark tasks complete in `tasks.md` AS you go, not at the end\n- Before returning, re-read the persisted tasks artifact and ensure completed tasks are visibly marked `[x]`; internal todos are not completion evidence\n- If you discover the design is wrong or incomplete, NOTE IT in your return summary — don't silently deviate\n- If a task is blocked by something unexpected, STOP and report back\n- If workload forecast requires a decision and none was provided, STOP before writing code\n- When applying a chained\u002Fstacked PR slice, keep the batch autonomous: one deliverable scope, verification included, and clear rollback boundary\n- When applying `size:exception`, state it explicitly in apply-progress and the return summary\n- NEVER implement tasks that weren't assigned to you\n- Skill loading is handled in Step 1 — follow any loaded skills strictly when writing code\n- Apply any `rules.apply` from `openspec\u002Fconfig.yaml`\n- If Strict TDD Mode is active (Step 3), load `strict-tdd.md` and follow its cycle INSTEAD of Step 4\n- When Strict TDD is active, the `strict-tdd.md` module's rules OVERRIDE Step 4 entirely\n- Return envelope per **Section D** from `skills\u002F_shared\u002Fsdd-phase-common.md`.\n",{"name":5,"license":15,"metadata":24,"description":7,"user-invocable":20,"disable-model-invocation":27},{"author":25,"version":26,"delegate_only":27},"gentleman-programming","3.0",true,"main","6745e47ba2ea20f71dce9a8303c332ef63ef43b6",".claude\u002Fskills\u002Fsdd-apply","https:\u002F\u002Fgithub.com\u002FAdrichDev\u002Fgru_orchestrator","https:\u002F\u002Fgithub.com\u002FAdrichDev","https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fadrichv\u002F","Software Junior Developer en continua formación y aprendizaje. Además , Trabajador social.\r\nIntentaré ir subiendo cositas!! \r\nFeliz día!!","2026-11-15 04:14:37",{"source":8,"skillName":5,"commands":37,"prompt":47,"zipAvailable":27},[38,41,44],{"tool":39,"command":40},"npx","npx skills add AdrichDev\u002Fgru_orchestrator --skill sdd-apply",{"tool":42,"command":43},"bunx","bunx skills add AdrichDev\u002Fgru_orchestrator --skill sdd-apply",{"tool":45,"command":46},"pnpm","pnpm dlx skills add AdrichDev\u002Fgru_orchestrator --skill sdd-apply","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FAdrichDev\u002Fgru_orchestrator\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fsdd-apply\u002FSKILL.md",[49],{"path":50,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":53,"sourceHtmlUrl":54},"strict-tdd.md","REFERENCE","参考文档",18533,"https:\u002F\u002Fgithub.com\u002FAdrichDev\u002Fgru_orchestrator\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fsdd-apply\u002Fstrict-tdd.md",[56,63,68,76,83,90],{"skillNo":57,"name":58,"nameZh":6,"description":59,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":60,"categories":61,"keywords":62,"publishedAt":19},"01M75ED2J1WS4B52JV2Q0R01CB","invocable-skill","Run a security audit on the specified target. Use when asked to audit, scan, or assess security posture.","https:\u002F\u002Fgithub.com\u002FAdrichDev\u002Fgru_orchestrator\u002Fblob\u002Fmain\u002Fpackages\u002Fkernel\u002Fsrc\u002Fadapters\u002F__tests__\u002Fmock-ac-root\u002Fskills\u002Finvocable-skill\u002FSKILL.md",[],[],{"skillNo":64,"name":58,"nameZh":6,"description":59,"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},"01M75ECRCQFK3WY7Q990S3ZANF","https:\u002F\u002Fgithub.com\u002FAdrichDev\u002Fgru_orchestrator\u002Fblob\u002Fmain\u002Fpackages\u002Fkernel\u002Fsrc\u002Fadapters\u002F__tests__\u002Ffixtures-ac\u002Finvocable-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":72,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":19},"01M75ECJTFT8NZXFX0M95XHBGJ","work-unit-commits","Plan commits as reviewable work units. Trigger: implementation, commit splitting, chained PRs, or keeping tests and docs with code.","Apache-2.0","https:\u002F\u002Fgithub.com\u002FAdrichDev\u002Fgru_orchestrator\u002Fblob\u002Fmain\u002F.qwen\u002Fskills\u002Fwork-unit-commits\u002FSKILL.md",[],[],{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":19},"01M75ECHC0CRV7MCB19WVAPQZN","skill-registry","Trigger: update skills, skill registry, actualizar skills, after skill changes. Index available skills by trigger and path.","https:\u002F\u002Fgithub.com\u002FAdrichDev\u002Fgru_orchestrator\u002Fblob\u002Fmain\u002F.qwen\u002Fskills\u002Fskill-registry\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":72,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":19},"01M75ECFSPSZCFMA3BK74SN2M7","skill-improver","Trigger: improve skills, audit skills, refactor skills, skill quality. Audit and upgrade existing LLM-first skills.","https:\u002F\u002Fgithub.com\u002FAdrichDev\u002Fgru_orchestrator\u002Fblob\u002Fmain\u002F.qwen\u002Fskills\u002Fskill-improver\u002FSKILL.md",[],[],{"skillNo":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":72,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":19},"01M75ECE874149EMN8ZEEMSAPM","skill-creator","Trigger: new skills, agent instructions, documenting AI usage patterns. Create LLM-first skills with valid frontmatter.","https:\u002F\u002Fgithub.com\u002FAdrichDev\u002Fgru_orchestrator\u002Fblob\u002Fmain\u002F.qwen\u002Fskills\u002Fskill-creator\u002FSKILL.md",[],[]]