@aisdlc/core 3.1.0 → 3.2.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/dist/index.cjs CHANGED
@@ -14210,6 +14210,7 @@ Use this skill when starting a **brand-new software project from scratch (Greenf
14210
14210
  3. **Release Consolidation (\`release/v1.0.0\` \u2794 \`main\`)**: Once all MVP \`CHG-*\` increments are integrated (\`sdd_integrate\`), run \`verify\` (\`npx aisdlc verify all\`), generate SBOM (\`npx aisdlc verify licenses --sbom reports/sbom.cdx.json --notices THIRD_PARTY_NOTICES.md\`), compute release KPIs (\`kpi_release\`), and open the Release PR into \`main\` for human Product Owner & Release Manager sign-off.
14211
14211
 
14212
14212
  ## Non-Negotiable Guardrails
14213
+ - **Clean Context Window Isolation per Specialist Agent (Cross-IDE)**: Never execute two distinct specialist roles in the same conversation context window. In subagent-capable IDEs (Antigravity \`invoke_subagent\`, Claude Code \`Task\` / \`.claude/agents/\`, GitHub Copilot \`@agent-*\`), always spawn each specialist in a fresh isolated subagent context that reads state strictly from disk (\`specs/\`, \`handoff.yaml\`, \`get_active_handoff\`). In chat-based IDEs (Cursor, Windsurf, Aider), persist all artifacts to disk, emit the Workflow Handoff block, and instruct the user to open a **New Chat** (or \`/clear\`) before running the next specialist.
14213
14214
  - **FORBIDDEN Greenfield Monolith (\`1 UC-* = 1 CHG-*\`)**: Never bundle the entire product or multiple \`UC-*\` into a single SDD change (\`CHG-001\`). Deterministic Gate 3 (\`verify_governance\`) blocks any \`handoff.yaml\` containing $> 1$ Use Case (\`Greenfield monolith detected\`). Parallelism is achieved by running **multiple bounded \`CHG-*\` concurrently** on separate branches/worktrees, never by stuffing multiple \`UC-*\` into one \`CHG-*\`.
14214
14215
  - Never invoke \`sdd_change_new\` or write application code in \`src/\` on a Greenfield project before \`ARCH-INTRO-001\`, \`agent-expert-user\` MVP validation, and at least one \`UC-*\` and \`FR-*\` exist.
14215
14216
  - All newly drafted specifications must start with \`status: draft\`.
@@ -14253,6 +14254,7 @@ Instead of attempting to document 100% of a legacy monolith upfront, apply **bou
14253
14254
  Once the target domain's baseline artifacts (\`ARCH-INTRO-*\`, \`UC-*\`, \`FR-*\`, \`CMP-*\`) are in place, execute all new features, refactorings, or bugfixes via [\`aisdlc-increment\`](../aisdlc-increment/SKILL.md).
14254
14255
 
14255
14256
  ## Non-Negotiable Guardrails
14257
+ - **Clean Context Window Isolation per Specialist Agent (Cross-IDE)**: Never execute two distinct specialist roles in the same conversation context window. In subagent-capable IDEs (Antigravity \`invoke_subagent\`, Claude Code \`Task\` / \`.claude/agents/\`, GitHub Copilot \`@agent-*\`), always spawn each specialist in a fresh isolated subagent context that reads state strictly from disk (\`specs/\`, \`handoff.yaml\`, \`get_active_handoff\`). In chat-based IDEs (Cursor, Windsurf, Aider), persist all artifacts to disk, emit the Workflow Handoff block, and instruct the user to open a **New Chat** (or \`/clear\`) before running the next specialist.
14256
14258
  - **Zero Drive-By Rewrites**: Never mass-refactor untouched legacy modules during onboarding. Apply the Strangler / Boy Scout rule: bring touched functions into compliance with \`quality-policy.yaml\` incrementally via governed SDD tasks (\`TSK-*\`).
14257
14259
  - **Characterization First**: Never alter legacy business logic without first capturing its current behavior in an \`FR-*\` Gherkin specification and a verifiable test suite (\`agent-qa-engineer\`).
14258
14260
  - **Immediate Secret Remediation**: If Step 2 (\`verify_security\`) uncovers leaked credentials (exit code 4), halt feature work and remediate/rotate secrets immediately.`
@@ -14321,6 +14323,7 @@ Use this skill for **every evolutionary increment (new feature, enhancement, ref
14321
14323
  - Open the Release PR from \`release/vX.Y.Z\` to \`main\` for final human Product Owner and Release Manager approval and semantic Git tagging (\`vX.Y.Z\`).
14322
14324
 
14323
14325
  ## Non-Negotiable Guardrails
14326
+ - **Clean Context Window Isolation per Specialist Agent (Cross-IDE)**: Never execute two distinct specialist roles (e.g., \`agent-qa-engineer\` and \`agent-developer\`, or \`agent-developer\` and \`agent-security-auditor\`) in the same conversation context window. In subagent-capable IDEs (Antigravity \`invoke_subagent\`, Claude Code \`Task\` / \`.claude/agents/\`, GitHub Copilot \`@agent-*\`), always spawn each specialist in a fresh isolated subagent context that reads state strictly from disk (\`specs/\`, \`handoff.yaml\`, \`get_active_handoff\`). In chat-based IDEs (Cursor, Windsurf, Aider), persist all artifacts to disk, emit the Workflow Handoff block, and instruct the user to open a **New Chat** (or \`/clear\`) before running the next specialist.
14324
14327
  - **Single-Use-Case Scope (\`1 UC-* = 1 CHG-*\`)**: Each SDD change (\`CHG-*\`) must deliver at most 1 Use Case (\`UC-*\`). Bundling multiple \`UC-*\` into one \`handoff.yaml\` is blocked by \`verify_governance\` (\`Greenfield monolith detected\`). Scale throughput via parallel \`CHG-*\` branches/worktrees, not monolithic changes.
14325
14328
  - **100% Design & Handoff Coverage in \`tasks.md\`**: Every \`CMP-*\`, \`FR-*\`, \`SEC-REQ-*\`, and \`PRIV-REQ-*\` declared in \`design.md\` and \`handoff.yaml\` must be explicitly assigned to at least one \`TSK-*\` in \`tasks.md\`.`
14326
14329
  }
@@ -14512,6 +14515,7 @@ This file defines the operational rules and specialized role mappings for Google
14512
14515
  ---
14513
14516
 
14514
14517
  ## 4. Workflow Handoff Protocol and Human Action Window | Ventana de Acci\xF3n Humana
14518
+ - **Clean Context Window Isolation per Specialist Agent (Cross-IDE)**: Never execute two distinct specialist roles in the same conversation context window. In subagent-capable IDEs (Antigravity \`invoke_subagent\`, Claude Code \`Task\` / \`.claude/agents/\`, GitHub Copilot \`@agent-*\`), always delegate each specialist phase to a fresh isolated subagent context that hydrates state strictly from disk (\`specs/\`, \`handoff.yaml\`, \`get_active_handoff\`). In chat-based IDEs (Cursor, Windsurf, Aider), persist all artifacts to disk, emit the Workflow Handoff block, and instruct the user to open a **New Chat** (or \`/clear\`) before running the next specialist.
14515
14519
  - **Conditional Activation Rule by Autonomy**:
14516
14520
  - \u{1F7E2} **\`AUTONOMOUS\`** (or exclusive final supervision in PR/CI): **OMITTED**. Do not request or emit interactive handoff to avoid interrupting unattended execution.
14517
14521
  - \u{1F7E1} **Autonomy $\\ge$ \`HUMAN_REVIEW_PLAN\`** (\`HUMAN_REVIEW_PLAN\`, \`AMBIGUOUS\`, \`HIGH_RISK_MANUAL\`): **MANDATORY**. Emit the Workflow Handoff block compliant with [\`templates/workflow/agent-handoff.template.md\`](../../templates/workflow/agent-handoff.template.md) and **STOP**.
@@ -14570,6 +14574,7 @@ Before finalizing changes or proposing commits:
14570
14574
  - \u{1F534} **\`HIGH_RISK_MANUAL\`**: Critical risk. Human-only execution.
14571
14575
 
14572
14576
  ## 6. Workflow Handoff Protocol and Human Action Window | Ventana de Acci\xF3n Humana
14577
+ - **Clean Context Window Isolation per Specialist Agent (Cross-IDE)**: Never execute two distinct specialist roles in the same conversation context window. In subagent-capable IDEs (Antigravity \`invoke_subagent\`, Claude Code \`Task\` / \`.claude/agents/\`, GitHub Copilot \`@agent-*\`), always delegate each specialist phase to a fresh isolated subagent context that hydrates state strictly from disk (\`specs/\`, \`handoff.yaml\`, \`get_active_handoff\`). In chat-based IDEs (Cursor, Windsurf, Aider), persist all artifacts to disk, emit the Workflow Handoff block, and instruct the user to open a **New Chat** (or \`/clear\`) before running the next specialist.
14573
14578
  - **Conditional Activation**:
14574
14579
  * \u{1F7E2} **\`AUTONOMOUS\`** (or PR/CI supervision): **OMITTED**. Do not request or emit interactive handoff.
14575
14580
  * \u{1F7E1} **Autonomy $\\ge$ \`HUMAN_REVIEW_PLAN\`** (\`HUMAN_REVIEW_PLAN\`, \`AMBIGUOUS\`, \`HIGH_RISK_MANUAL\`): **MANDATORY**. Emit Workflow Handoff block compliant with [\`templates/workflow/agent-handoff.template.md\`](templates/workflow/agent-handoff.template.md) and **STOP**.
@@ -14774,6 +14779,7 @@ Change-ID: CHG-026-AGENT-NATIVE-CONFIGS
14774
14779
  ---
14775
14780
 
14776
14781
  ## 7. Workflow Handoff Protocol and Human Action Window | Ventana de Acci\xF3n Humana
14782
+ - **Clean Context Window Isolation per Specialist Agent (Cross-IDE)**: Never execute two distinct specialist roles in the same conversation context window. In subagent-capable IDEs (Antigravity \`invoke_subagent\`, Claude Code \`Task\` / \`.claude/agents/\`, GitHub Copilot \`@agent-*\`), always delegate each specialist phase to a fresh isolated subagent context that hydrates state strictly from disk (\`specs/\`, \`handoff.yaml\`, \`get_active_handoff\`). In chat-based IDEs (Cursor, Windsurf, Aider), persist all artifacts to disk, emit the Workflow Handoff block, and instruct the user to open a **New Chat** (or \`/clear\`) before running the next specialist.
14777
14783
  - **Conditional Activation**:
14778
14784
  - \u{1F7E2} **\`AUTONOMOUS\`** (or PR/CI final supervision): **OMITTED**. Uninterrupted continuous execution.
14779
14785
  - \u{1F7E1} **Autonomy $\\ge$ \`HUMAN_REVIEW_PLAN\`** (\`HUMAN_REVIEW_PLAN\`, \`AMBIGUOUS\`, \`HIGH_RISK_MANUAL\`): **MANDATORY**. Emit Workflow Handoff block ([\`templates/workflow/agent-handoff.template.md\`](templates/workflow/agent-handoff.template.md)) and **STOP**.
@@ -14841,6 +14847,7 @@ Non-Negotiable Quality Thresholds (\`quality-policy.yaml\`):
14841
14847
  ---
14842
14848
 
14843
14849
  ## 5. Workflow Handoff Protocol and Human Action Window | Ventana de Acci\xF3n Humana
14850
+ - **Clean Context Window Isolation per Specialist Agent (Cross-IDE)**: Never execute two distinct specialist roles in the same conversation context window. In subagent-capable IDEs (Antigravity \`invoke_subagent\`, Claude Code \`Task\` / \`.claude/agents/\`, GitHub Copilot \`@agent-*\`), always delegate each specialist phase to a fresh isolated subagent context that hydrates state strictly from disk (\`specs/\`, \`handoff.yaml\`, \`get_active_handoff\`). In chat-based IDEs (Cursor, Windsurf, Aider), persist all artifacts to disk, emit the Workflow Handoff block, and instruct the user to open a **New Chat** (or \`/clear\`) before running the next specialist.
14844
14851
  - **Conditional Activation**:
14845
14852
  * \u{1F7E2} **\`AUTONOMOUS\` Mode** (or PR/CI final supervision): **OMITTED**. Do not interrupt unattended execution.
14846
14853
  * \u{1F7E1} **Autonomy $\\ge$ \`HUMAN_REVIEW_PLAN\`** (\`HUMAN_REVIEW_PLAN\`, \`AMBIGUOUS\`, \`HIGH_RISK_MANUAL\`): **MANDATORY**. Emit Workflow Handoff block compliant with [\`templates/workflow/agent-handoff.template.md\`](../templates/workflow/agent-handoff.template.md) and **STOP**.
@@ -15981,7 +15988,7 @@ Describes the adversary's profile, technical skill level, available resources, a
15981
15988
  | :--- | :--- | :--- | :--- | :--- |\r
15982
15989
  | **1.0.0** | 2026-09-03 | Security Officer / Threat Modeler | Initial threat profile characterization | CHG-SEC-001 |\r
15983
15990
  `;
15984
- var STARTER_TEMPLATES_WORKFLOW_AGENT_HANDOFF_TEMPLATE_MD = '# Workflow Handoff Template\r\n\r\n> **Canonical reference:** [`process/09_agent_protocols.md`](../../process/09_agent_protocols.md) and [`process/01_governance_and_roles.md`](../../process/01_governance_and_roles.md).\r\n>\r\n> **Activation Criterion by Autonomy Mode:**\r\n> - \u{1F7E2} **`AUTONOMOUS` (or exclusive supervision at the end in PR/CI):** OMITTED. The agent executes and delivers without interrupting the user with this handoff block.\r\n> - \u{1F7E1} **`HUMAN_REVIEW_PLAN` | \u{1F7E0} `AMBIGUOUS` | \u{1F534} `HIGH_RISK_MANUAL`:** MANDATORY. Upon completing the phase or proposal, the agent emits this block and **halts**, granting visibility to the user and relinquishing control.\r\n\r\n---\r\n\r\n### \u{1F504} Workflow Handoff\r\n\r\n- **Completed Phase / Activity**: `[Current phase, e.g. Product Definition / Threat Modeling / Privacy Modeling / Architecture / SDD Implementation]`\r\n- **Acting Role**: `[agent-product-analyst | agent-threat-modeler | agent-privacy-officer | agent-system-architect | agent-developer | agent-expert-user | agent-security-auditor | agent-compliance-checker]`\r\n- **Task Autonomy Mode**: `[HUMAN_REVIEW_PLAN | AMBIGUOUS | HIGH_RISK_MANUAL]`\r\n- **Produced Deliverables**:\r\n - [x] `[Path or ID of generated artifact, e.g. specs/product/uc-mission.md]`\r\n - [x] `[Path or ID of generated artifact, e.g. quality-report.md]`\r\n\r\n- **Recommended Next Role(s) in Workflow**:\r\n - `[Suggested role name, e.g. agent-threat-modeler]`: `[Technical rationale, e.g. Model STRIDE adversaries and mitigate abuse cases]`\r\n - *(Optional)* `[Alternative role, e.g. agent-expert-user]`: `[Technical rationale, e.g. Evaluate UX and MVP boundary]`\r\n\r\n- **Suggested Invocation Prompt**:\r\n > "Act as `[next-role]` and proceed with the `[next phase]` phase on the generated deliverables: `[key artifacts]`."\r\n\r\n- **Human Action Window (User Sovereignty and Priority)**:\r\n - \u270B **Review and Validate**: Inspect generated artifacts and validate alignment with project vision before proceeding.\r\n - \u270F\uFE0F **Edit Directly**: Modify requirements, code, or acceptance criteria directly at any time.\r\n - \u{1F6D1} **Pause, Divert, or Discard**: Freeze workflow, redirect toward another goal, or discard proposal.\r\n - \u25B6\uFE0F **Delegate to Next Agent**: If satisfied with output, execute suggested prompt to continue the chain.\r\n\r\n---\r\n';
15991
+ var STARTER_TEMPLATES_WORKFLOW_AGENT_HANDOFF_TEMPLATE_MD = '# Workflow Handoff Template\r\n\r\n> **Canonical reference:** [`process/09_agent_protocols.md`](../../process/09_agent_protocols.md) and [`process/01_governance_and_roles.md`](../../process/01_governance_and_roles.md).\r\n>\r\n> **Activation Criterion by Autonomy Mode:**\r\n> - \u{1F7E2} **`AUTONOMOUS` (or exclusive supervision at the end in PR/CI):** OMITTED. The agent executes and delivers without interrupting the user with this handoff block.\r\n> - \u{1F7E1} **`HUMAN_REVIEW_PLAN` | \u{1F7E0} `AMBIGUOUS` | \u{1F534} `HIGH_RISK_MANUAL`:** MANDATORY. Upon completing the phase or proposal, the agent emits this block and **halts**, granting visibility to the user and relinquishing control.\r\n\r\n---\r\n\r\n### \u{1F504} Workflow Handoff\r\n\r\n- **Completed Phase / Activity**: `[Current phase, e.g. Product Definition / Threat Modeling / Privacy Modeling / Architecture / SDD Implementation]`\r\n- **Acting Role**: `[agent-product-analyst | agent-threat-modeler | agent-privacy-officer | agent-system-architect | agent-developer | agent-expert-user | agent-security-auditor | agent-compliance-checker]`\r\n- **Task Autonomy Mode**: `[HUMAN_REVIEW_PLAN | AMBIGUOUS | HIGH_RISK_MANUAL]`\r\n- **Produced Deliverables**:\r\n - [x] `[Path or ID of generated artifact, e.g. specs/product/uc-mission.md]`\r\n - [x] `[Path or ID of generated artifact, e.g. quality-report.md]`\r\n\r\n- **Recommended Next Role(s) in Workflow**:\r\n - `[Suggested role name, e.g. agent-threat-modeler]`: `[Technical rationale, e.g. Model STRIDE adversaries and mitigate abuse cases]`\r\n - *(Optional)* `[Alternative role, e.g. agent-expert-user]`: `[Technical rationale, e.g. Evaluate UX and MVP boundary]`\r\n\r\n- **Suggested Invocation Prompt**:\r\n > "Act as `[next-role]` and proceed with the `[next phase]` phase on the generated deliverables: `[key artifacts]`."\r\n- **Clean Context Window Execution (Cross-IDE)**:\r\n - \u{1F9F9} **Subagent-Capable IDEs (Antigravity / Claude Code / Copilot)**: Delegate automatically via subagent tool (`invoke_subagent` / `Task` / `@agent-*`) to launch a fresh context window.\r\n - \u{1F195} **Chat-Based IDEs (Cursor / Windsurf / Aider)**: Open a **New Chat (`Cmd/Ctrl + N`)** or run `/clear` before pasting the Suggested Invocation Prompt so the next specialist starts with zero context pollution and hydrates state exclusively from `handoff.yaml` / `get_active_handoff`.\r\n\r\n- **Human Action Window (User Sovereignty and Priority)**:\r\n - \u270B **Review and Validate**: Inspect generated artifacts and validate alignment with project vision before proceeding.\r\n - \u270F\uFE0F **Edit Directly**: Modify requirements, code, or acceptance criteria directly at any time.\r\n - \u{1F6D1} **Pause, Divert, or Discard**: Freeze workflow, redirect toward another goal, or discard proposal.\r\n - \u25B6\uFE0F **Delegate to Next Agent**: If satisfied with output, execute suggested prompt to continue the chain.\r\n\r\n---\r\n';
15985
15992
  var STARTER__GITHUB_PULL_REQUEST_TEMPLATE_MD = "## 1. SDD Identification and Traceability\r\n\r\n- **SDD Change Specification**: `specs/changes/active/<change-id>/` <!-- Link to active change directory -->\r\n- **PDaC Handoff Sidecar**: `specs/changes/active/<change-id>/handoff.yaml` <!-- ID: HOF-* -->\r\n- **Git Hierarchy Tier (4 Tiers)**:\r\n - [ ] Tier 4: `task/<PARENT-ID>/<TSK-ID>-<slug>` \u2794 `feat/<PARENT-ID>-<slug>`\r\n - [ ] Tier 3: `feat/<FEAT-ID>-<slug>` \u2794 `release/vX.Y.Z`\r\n - [ ] Tier 2: `release/vX.Y.Z` \u2794 `main`\r\n- **Linked Issues**: Closes #... / Refs #...\r\n\r\n---\r\n\r\n## 2. \u{1F4CB} Plan Execution Matrix and Testing Status (`X` vs `O`)\r\n\r\n> [!IMPORTANT]\r\n> **Non-Negotiable Total Exhaustiveness Rule**:\r\n> This table must include **each and every item identified for implementation** in the delivery plan (originating from `tasks.md`, the SDD specification, or issue criteria), without omitting any or creating abstract groupings.\r\n>\r\n> **Mandatory Marking Convention**:\r\n> - **`[X]` (Completed and Tested)**: Item is 100% implemented and has verifiable automated tests on disk (`tests/` or `.feature`) passing green.\r\n> - **`[O]` (Problem / Blocked)**: If there was any issue, technical limitation, failing test, or deferred scope, it must be marked with `[O]` and **full mandatory justification is required**:\r\n> 1. **Technical root cause** or encountered limitation.\r\n> 2. **Impact** on increment or system.\r\n> 3. **Mitigation** or rationale for deferral.\r\n> 4. **Formal reference** to tracking issue (GitHub Issue or `ADR-TECH-DEBT-*`).\r\n>\r\n> *Any PR with omitted planned items or unjustified `[O]` will be immediately blocked and rejected by the Tech Lead.*\r\n\r\n| Task / Item ID | Planned Description | Status (`[X]` / `[O]`) | Test Evidence on Disk or Mandatory Justification of `[O]` |\r\n| :--- | :--- | :---: | :--- |\r\n| `TSK-001` | <!-- e.g. Contract DTOs and Interfaces Definition --> | **[X]** | Covered in `tests/unit/dto.spec.ts` (100% pass) |\r\n| `TSK-002` | <!-- e.g. Network disconnection resilience --> | **[O]** | **Root cause:** TCP socket mock exhibits race conditions in CI.<br>**Impact:** Low in prod, flaky test.<br>**Mitigation:** Temporary extended timeout; tracked in issue #... |\r\n\r\n---\r\n\r\n## 3. \u{1F5FA}\uFE0F Weak Points Map and Risk Hotspots (Weak Points Hotspots)\r\n\r\n### A. Complexity Hotspots\r\n<!-- Record functions nearing quality-policy.yaml thresholds: CC > 7 (max 10), Cognitive > 10 (max 15), LOC > 30 (max 40), MI < 70 (min 50/65) -->\r\n| File | Function / Module | Metrics (CC / Cognitive / LOC / MI) | Technical Maintainability Justification |\r\n| :--- | :--- | :---: | :--- |\r\n| <!-- e.g. src/engine.ts --> | <!-- parsePayload() --> | CC: 8, Cog: 11, LOC: 32, MI: 72 | Deterministic parser without recursion; exhaustively tested. |\r\n\r\n### B. Edge Cases and Blind Spots\r\n<!-- Scenarios not covered by automated unit tests (concurrency, network timeouts, memory, jitter) -->\r\n- [ ] **Edge Case 1**: <!-- Scenario description, estimated risk, and operational mitigation -->\r\n- [ ] **Edge Case 2**: <!-- Scenario description, estimated risk, and operational mitigation -->\r\n\r\n### C. AI Assumptions\r\n<!-- Behaviors or heuristics assumed by the agent during implementation that require critical human review -->\r\n- [ ] **Assumption 1**: <!-- e.g. Assumed payload always arrives pre-parsed in UTF-8 -->\r\n- [ ] **Assumption 2**: <!-- e.g. Default timeout set to 5000ms as not stipulated in spec -->\r\n\r\n---\r\n\r\n## 4. \u{1F6E1}\uFE0F Attack Surface and Cybersecurity (STRIDE / Zero Trust)\r\n\r\n- [ ] **New External Channels**: Does it introduce HTTP endpoints, sockets, IPC channels, or brokers?\r\n- [ ] **Sanitization and Inputs**: Are all external inputs validated with strict typing schemas (Zod/JSON Schema)?\r\n- [ ] **Insecure Serialization**: Is polymorphic deserialization or dynamic code evaluation avoided?\r\n- [ ] **Secrets and Enclaves**: Are cryptographic keys, credentials, sensitive env vars, or enclaves (`SEC-ENC-*`) handled?\r\n- [ ] **Mitigated Security Requirements**: Implemented IDs (e.g. `SEC-REQ-*`) and associated tests (`SEC-TEST-*` or `.feature`).\r\n\r\n---\r\n\r\n## 5. \u{1F4E6} Dependencies and Licenses Inspection\r\n\r\n- **New Dependencies Added**:\r\n | Package | Version | License Type | `license-policy.yaml` Verification |\r\n | :--- | :---: | :---: | :---: |\r\n | <!-- e.g. zod --> | <!-- ^3.22.4 --> | <!-- MIT --> | <!-- \u2714 Compliant (Permissive) --> |\r\n- [ ] Zero dependencies with prohibited licenses (GPLv3/AGPL/unapproved commercial).\r\n- [ ] Verification executed successfully (`pnpm run verify:licenses`).\r\n\r\n---\r\n\r\n## 6. \u2705 Deterministic Pre-Flight Checklist (Release Gates)\r\n\r\n- [ ] **Strict Typing and Linter**: `pnpm run typecheck` (zero errors, zero unjustified suppressions).\r\n- [ ] **Quality Gate**: `pnpm run verify:quality` (CC $\\le 10$, Cognitive $\\le 15$, MI $\\ge 50$, LOC $\\le 40$).\r\n- [ ] **BDD / Unit Tests**: `pnpm run verify:testing` (Gherkin scenarios synchronized and 100% passing).\r\n- [ ] **360\xB0 Traceability**: `pnpm run verify:traceability` (no orphan requirements).\r\n- [ ] **Open Source Licenses**: `pnpm run verify:licenses` (strict policy compliance).\r\n- [ ] **Schema Compliance**: `pnpm run verify:schemas` (100% of artifacts conform to JSON Schema).\r\n- [ ] **Full Test Suite**: `pnpm run test:all` / `pnpm run verify:all` (general green CI approval).\r\n\r\n---\r\n\r\n## 7. \u{1F6A8} Contingency Plan, Observability, and Rollback\r\n\r\n- **Observability Metrics and Alarms**: <!-- What metrics (p99 latency, 5xx error rate, memory) or logs should be monitored post-deployment? -->\r\n- **Rollback Procedure**: <!-- Exact command or procedure to revert change without collateral impact (e.g. git revert -m 1 <sha>) -->\r\n\r\n---\r\n\r\n## 8. \u{1F4CA} AI-SDLC: Aggregated KPIs Summary\r\n\r\n<!-- AI-SDLC-KPI-SUMMARY-START -->\r\n| Metric | Human | Total PR |\r\n| :--- | :---: | :---: |\r\n| **Commits Made** | - | **-** |\r\n| **Lines Added / Modif.** | - | **-** |\r\n| **Active Development Time** | - | **-** |\r\n| **Consumed Tokens (In+Out)** | - | **-** |\r\n| **Estimated Cost (\u20AC/$)** | \u2014 (Labor) | **-** |\r\n\r\n> *This block is updated automatically via `aisdlc kpi pr` or the CI workflow.*\r\n<!-- AI-SDLC-KPI-SUMMARY-END -->\r\n";
15986
15993
  var STARTER_COMPLIANCE_LICENSE_MANIFEST_YAML = '# ==============================================================================\r\n# AI-SDLC: Manifiesto de Cumplimiento de Licencias Evaluadas\r\n# ==============================================================================\r\n# Verificado contra: /license-policy.yaml\r\n# ==============================================================================\r\n\r\nevaluated_dependencies:\r\n - package: "@modelcontextprotocol/sdk"\r\n version: "1.0.4"\r\n spdx_license: "MIT"\r\n category: "permissive_free"\r\n status: "APPROVED"\r\n usage: "SDK oficial para servidor Model Context Protocol."\r\n commercial_payment_required: false\r\n\r\n - package: "commander"\r\n version: "12.1.0"\r\n spdx_license: "MIT"\r\n category: "permissive_free"\r\n status: "APPROVED"\r\n usage: "Parser de l\xEDnea de comandos para @ai-sdlc/cli."\r\n commercial_payment_required: false\r\n\r\n - package: "zod"\r\n version: "3.23.8"\r\n spdx_license: "MIT"\r\n category: "permissive_free"\r\n status: "APPROVED"\r\n usage: "Validaci\xF3n de esquemas y tipos en runtime."\r\n commercial_payment_required: false\r\n\r\n - package: "js-yaml"\r\n version: "4.1.0"\r\n spdx_license: "MIT"\r\n category: "permissive_free"\r\n status: "APPROVED"\r\n usage: "Parser y serializador YAML para especificaciones y frontmatter."\r\n commercial_payment_required: false\r\n\r\n - package: "chalk"\r\n version: "5.3.0"\r\n spdx_license: "MIT"\r\n category: "permissive_free"\r\n status: "APPROVED"\r\n usage: "Estilizado de salida en terminal para comandos CLI."\r\n commercial_payment_required: false\r\n\r\n - package: "ajv"\r\n version: "8.17.1"\r\n spdx_license: "MIT"\r\n category: "permissive_free"\r\n status: "APPROVED"\r\n usage: "Validador JSON Schema para ProductShape, arc42 y SDD."\r\n commercial_payment_required: false\r\n';
15987
15994
  var STARTER_PROCESS_README = "# AI-SDLC: Process & Governance Framework\n\nThis directory hosts the deterministic governance, verification gates, and lifecycle protocols for AI-assisted software development.\n\n## Core Protocols & Lifecycle\n- **Phase 0 (Product Definition)**: `specs/product/` (Requirements `FR-*`, Use Cases `UC-*`, Actors `ACT-*`, Rules `BR-*`).\n- **Phase 1 (System Architecture)**: `templates/architecture/` (arc42 + NAF v4, `CMP-*`, Decisions `ADR-*`).\n- **Phase 2 (Cybersecurity Triad)**: `templates/security/` (Threat Actors `ACT-THREAT-*`, Abuse Cases `ABUSE-*`, Security Reqs `SEC-REQ-*`).\n- **Phase 2b (Privacy-by-Design as Code)**: `templates/privacy/` (Data Assets `DATA-ASSET-*`, ROPA `PRIV-ROPA-*`, Privacy Reqs `PRIV-REQ-*`).\n- **Phase 3 (Spec-Driven Delivery - SDD)**: `specs/changes/active/<change-id>/` (`proposal.md`, `spec.md`, `design.md`, `tasks.md`, `handoff.yaml`).\n- **Phase 4 (Continuous Quality Gates & Verification)**:\n - Quality Gate: `npx aisdlc verify quality` (CC $\\le 10$, MI $\\ge 50$, LOC $\\le 40$)\n - BDD Testing: `npx aisdlc verify testing`\n - Traceability: `npx aisdlc verify traceability`\n - Licenses: `npx aisdlc verify licenses`\n - Privacy & Data Protection: `npx aisdlc verify privacy`\n - Schemas: `npx aisdlc verify schemas`\n - Release Gates: `npx aisdlc check` / `npx aisdlc verify all`\n";
@@ -16243,6 +16250,9 @@ var STARTER_WORKFLOW_HANDOFF_TEMPLATE = `# Workflow Handoff Template
16243
16250
 
16244
16251
  - **Suggested Invocation Prompt**:
16245
16252
  > "Act as \`[next-role]\` and proceed with the \`[next phase]\` phase on the generated deliverables: \`[key artifacts]\`."
16253
+ - **Clean Context Window Execution (Cross-IDE)**:
16254
+ - \u{1F9F9} **Subagent-Capable IDEs (Antigravity / Claude Code / Copilot)**: Delegate automatically via subagent tool (\`invoke_subagent\` / \`Task\` / \`@agent-*\`) to launch a fresh context window.
16255
+ - \u{1F195} **Chat-Based IDEs (Cursor / Windsurf / Aider)**: Open a **New Chat (\`Cmd/Ctrl + N\`)** or run \`/clear\` before pasting the Suggested Invocation Prompt so the next specialist starts with zero context pollution and hydrates state exclusively from \`handoff.yaml\` / \`get_active_handoff\`.
16246
16256
 
16247
16257
  - **Human Action Window (User Sovereignty and Priority)**:
16248
16258
  - \u270B **Review and Validate**: Inspect generated artifacts and validate alignment with project vision before proceeding.