@vpxa/aikit 0.1.1 → 0.1.3

package/scaffold/definitions/bodies.mjs

@@ -117,7 +117,7 @@ Batch 2 (after batch 1):
 2. **Break tasks small** — 1-3 files per dispatch, clear scope, clear acceptance criteria
 3. **Maximize parallelism** — independent tasks MUST run as parallel \`runSubagent\` calls in the SAME function block. Sequential dispatch of parallelizable tasks is a protocol violation.
 4. **Fresh context per subagent** — paste relevant code, don't reference conversation history
-5. **Search KB before planning** — check past decisions with \`search()\`
+5. **Search AI Kit before planning** — check past decisions with \`search()\`
 6. **Route correctly** — brainstorming for design, decision protocol for tech choices
 7. **Never proceed without user approval** at 🛑 stops
 8. **Max 2 retries** then escalate to user
@@ -182,7 +182,7 @@ At session start, check for an active flow:
 | \`flow_status\` | Check current execution state |
 | \`flow_reset\` | Clear flow state to start over |`,
 
-  Planner: `**Read \`AGENTS.md\`** in the workspace root for project conventions and KB protocol.
+  Planner: `**Read \`AGENTS.md\`** in the workspace root for project conventions and AI Kit protocol.
 
 **Read _shared/code-agent-base.md NOW** — it contains the Information Lookup Order, FORGE, and handoff protocols.
 
@@ -199,7 +199,7 @@ At session start, check for an active flow:
 
 ## Planning Workflow
 
-1. **KB Recall** — Search for past plans, architecture decisions, known patterns. Check \`list()\` for stored knowledge.
+1. **AI Kit Recall** — Search for past plans, architecture decisions, known patterns. Check \`list()\` for stored knowledge.
 2. **FORGE Classify** — \`forge_classify({ task, files, root_path: "." })\` to determine complexity tier
 3. **FORGE Ground** — \`forge_ground\` to scope map, seed unknowns, load constraints
 4. **Research** — Delegate to Explorer and Researcher agents to gather context
@@ -252,7 +252,7 @@ At session start, check for an active flow:
 | \`adr-skill\` | When the plan involves non-trivial technical decisions — create executable ADRs |
 | \`session-handoff\` | When context window is filling up, planning session ending, or major milestone completed |`,
 
-  Implementer: `**Read \`AGENTS.md\`** in the workspace root for project conventions and KB protocol.
+  Implementer: `**Read \`AGENTS.md\`** in the workspace root for project conventions and AI Kit protocol.
 
 **Read _shared/code-agent-base.md NOW** — it contains the Information Lookup Order, FORGE, and handoff protocols.
 
@@ -269,13 +269,13 @@ At session start, check for an active flow:
 
 - **Test-first always** — No implementation without a failing test
 - **Minimal code** — Don't build what isn't asked for
-- **Follow existing patterns** — Search KB for conventions before creating new ones
+- **Follow existing patterns** — Search AI Kit for conventions before creating new ones
 - **Never modify tests to make them pass** — Fix the implementation instead
 - **Run \`check\` after every change** — Catch errors early
 - **Loop-break** — If the same test fails 3 times with the same error after your fixes, STOP. Re-read the error from scratch, check your assumptions with \`trace\` or \`symbol\`, and try a fundamentally different approach. Do not attempt a 4th fix in the same direction
 - **Think-first for complex tasks** — If a task involves 3+ files or non-obvious logic, outline your approach before writing code. Check existing patterns with \`search\` first. Design, then implement`,
 
-  Frontend: `**Read \`AGENTS.md\`** in the workspace root for project conventions and KB protocol.
+  Frontend: `**Read \`AGENTS.md\`** in the workspace root for project conventions and AI Kit protocol.
 
 **Read _shared/code-agent-base.md NOW** — it contains the Information Lookup Order, FORGE, and handoff protocols.
 
@@ -294,13 +294,13 @@ At session start, check for an active flow:
 - **Responsive by default** — Mobile-first, test all breakpoints
 - **Test-first** — Component tests before implementation`,
 
-  Debugger: `**Read \`AGENTS.md\`** in the workspace root for project conventions and KB protocol.
+  Debugger: `**Read \`AGENTS.md\`** in the workspace root for project conventions and AI Kit protocol.
 
 **Read _shared/code-agent-base.md NOW** — it contains the Information Lookup Order, FORGE, and handoff protocols.
 
 ## Debugging Protocol
 
-1. **KB Recall** — Search for known issues matching this error pattern
+1. **AI Kit Recall** — Search for known issues matching this error pattern
 2. **Reproduce** — Confirm the error, use \`parse_output\` on stack traces and build errors for structured analysis
 3. **Verify targets exist** — Before tracing, confirm the files and functions mentioned in the error actually exist. Use \`find\` or \`symbol\` to verify paths and signatures. **Never trace into a file you haven't confirmed exists**
 4. **Trace** — \`symbol\`, \`trace\`, follow call chains backwards
@@ -318,13 +318,13 @@ At session start, check for an active flow:
 - **Verify before asserting** — Don't claim a function has a certain signature without checking via \`symbol\`. Don't reference a config option without confirming it exists in the codebase
 - **Break debug loops** — If you apply a fix, test, and get the same error 3 times: your hypothesis is wrong. STOP, discard your current theory, re-examine the error output and trace from a different entry point. Return \`ESCALATE\` if a fresh approach also fails`,
 
-  Refactor: `**Read \`AGENTS.md\`** in the workspace root for project conventions and KB protocol.
+  Refactor: `**Read \`AGENTS.md\`** in the workspace root for project conventions and AI Kit protocol.
 
 **Read _shared/code-agent-base.md NOW** — it contains the Information Lookup Order, FORGE, and handoff protocols.
 
 ## Refactoring Protocol
 
-1. **KB Recall** — Search for established patterns and conventions
+1. **AI Kit Recall** — Search for established patterns and conventions
 2. **Analyze** — \`analyze_structure\`, \`analyze_patterns\`, \`dead_symbols\`
 3. **Ensure test coverage** — Run existing tests, add coverage for untested paths
 4. **Refactor in small steps** — Each step must keep tests green
@@ -344,7 +344,7 @@ At session start, check for an active flow:
 |-------|--------------|
 | \`lesson-learned\` | After completing a refactor — extract principles from the before/after diff |`,
 
-  Security: `**Read \`AGENTS.md\`** in the workspace root for project conventions and KB protocol.
+  Security: `**Read \`AGENTS.md\`** in the workspace root for project conventions and AI Kit protocol.
 
 **Read _shared/code-agent-base.md NOW** — it contains the Information Lookup Order, FORGE, and handoff protocols.
 
@@ -359,7 +359,7 @@ At session start, check for an active flow:
 
 ## Security Review Protocol
 
-1. **KB Recall** — \`search("security findings <area>")\` + \`list()\` for past security decisions and known issues
+1. **AI Kit Recall** — \`search("security findings <area>")\` + \`list()\` for past security decisions and known issues
 2. **Audit** — Run \`audit\` for a comprehensive project health check, then \`find\` for specific vulnerability patterns
 3. **OWASP Top 10 Scan** — Check each category systematically
 4. **Dependency Audit** — Check for known CVEs in dependencies
@@ -389,7 +389,7 @@ At session start, check for an active flow:
 1. **[SEVERITY]** Title — Description, file:line, remediation
 \`\`\``,
 
-  Documenter: `**Read \`AGENTS.md\`** in the workspace root for project conventions and KB protocol.
+  Documenter: `**Read \`AGENTS.md\`** in the workspace root for project conventions and AI Kit protocol.
 
 **Read _shared/code-agent-base.md NOW** — it contains the Information Lookup Order, FORGE, and handoff protocols.
 
@@ -404,7 +404,7 @@ At session start, check for an active flow:
 
 ## Documentation Protocol
 
-1. **KB Recall** — \`search("documentation <area>")\` + \`list()\` for existing docs, conventions, architecture decisions
+1. **AI Kit Recall** — \`search("documentation <area>")\` + \`list()\` for existing docs, conventions, architecture decisions
 2. **Analyze** — \`analyze_structure\`, \`analyze_entry_points\`, \`file_summary\`
 3. **Draft** — Write documentation following project conventions
 4. **Cross-reference** — Link to related docs, ensure consistency
@@ -432,7 +432,7 @@ At session start, check for an active flow:
 | \`c4-architecture\` | When documenting system architecture — generate C4 Mermaid diagrams |
 | \`adr-skill\` | When documenting architecture decisions — create or update ADRs |`,
 
-  Explorer: `**Read \`AGENTS.md\`** in the workspace root for project conventions and KB protocol.
+  Explorer: `**Read \`AGENTS.md\`** in the workspace root for project conventions and AI Kit protocol.
 
 ## MANDATORY FIRST ACTION
 
@@ -448,7 +448,7 @@ At session start, check for an active flow:
 
 ## Exploration Protocol
 
-1. **KB Recall** — \`search\` for existing analysis on this area
+1. **AI Kit Recall** — \`search\` for existing analysis on this area
 2. **Discover** — Use \`find\`, \`symbol\`, \`scope_map\` to locate relevant files
 3. **Analyze** — Use \`analyze_structure\`, \`analyze_dependencies\`, \`file_summary\`
 4. **Compress** — Use \`compact\` for targeted file sections, \`digest\` when synthesizing 3+ sources, \`stratum_card\` for files you'll reference repeatedly

package/scaffold/definitions/plugins.mjs

@@ -11,7 +11,7 @@
 
 export const PLUGINS = {
   aikit: {
-    description: 'KB search, analysis, memory, and developer tools',
+    description: 'AI Kit search, analysis, memory, and developer tools',
     source: 'scaffold/general/skills/aikit/SKILL.md',
     required: true,
   },

package/scaffold/definitions/prompts.mjs

@@ -10,17 +10,17 @@
 
 export const PROMPTS = {
   ask: {
-    description: 'Quick research question — find answer using KB + web search',
+    description: 'Quick research question — find answer using AI Kit + web search',
     agent: 'Researcher-Alpha',
     tools: ['search', 'web_search', 'web_fetch', 'compact', 'file_summary', 'remember'],
     content: `## Quick Research
 
-1. **KB Recall** — Search knowledge base for existing answers
-2. **Web Search** — If KB has no relevant results, search the web
+1. **AI Kit Recall** — Search knowledge base for existing answers
+2. **Web Search** — If AI Kit has no relevant results, search the web
 3. **Synthesize** — Combine findings into a clear, concise answer
 4. **Remember** — If the answer is broadly useful, persist it to KB
 
-**Always cite sources** — KB entries, file paths, or URLs.`,
+**Always cite sources** — AI Kit entries, file paths, or URLs.`,
   },
 
   debug: {
@@ -40,7 +40,7 @@ export const PROMPTS = {
     content: `## Debug Workflow
 
 1. **Parse Error** — Use \`parse_output\` on the error message/stack trace
-2. **KB Recall** — Search for known issues matching this pattern
+2. **AI Kit Recall** — Search for known issues matching this pattern
 3. **Reproduce** — Run the failing command/test to confirm
 4. **Trace** — Use \`symbol\` and \`trace\` to follow the call chain
 5. **Diagnose** — Form hypothesis, gather evidence
@@ -91,7 +91,7 @@ Refer to the Orchestrator agent's full instructions for the detailed workflow.`,
     ],
     content: `## Planning Workflow
 
-1. **KB Recall** — Search for past plans, architecture decisions
+1. **AI Kit Recall** — Search for past plans, architecture decisions
 2. **FORGE Ground** — Classify tier, scope map, seed unknowns
 3. **Research** — Explore codebase, understand subsystems
 4. **Draft Plan** — 3-10 phases with agent assignments, TDD steps

package/scaffold/definitions/protocols.mjs

@@ -14,12 +14,12 @@ export const PROTOCOLS = {
 ## Invocation Mode Detection
 
 You may be invoked in two modes:
-1. **Direct** — you have full KB tool access. Follow the **Information Lookup Order** below.
+1. **Direct** — you have full AI Kit tool access. Follow the **Information Lookup Order** below.
 2. **Sub-agent** (via Orchestrator) — you may have limited MCP tool access.
-   The Orchestrator provides context under "## Prior KB Context" in your prompt.
-   If present, skip KB Recall and use the provided context instead.
+   The Orchestrator provides context under "## Prior AI Kit Context" in your prompt.
+   If present, skip AI Kit Recall and use the provided context instead.
 
-**Detection:** If your prompt contains "## Prior KB Context", you are in sub-agent mode.
+**Detection:** If your prompt contains "## Prior AI Kit Context", you are in sub-agent mode.
 
 ---
 
@@ -82,7 +82,7 @@ list()                     // see all stored knowledge entries
 
 ### Step 4: Tool Discovery
 
-If unsure which KB tool to use → run \`guide({ topic: "what you need" })\` for recommendations.
+If unsure which AI Kit tool to use → run \`guide({ topic: "what you need" })\` for recommendations.
 
 ## FORGE Protocol (Quality Gate)
 
@@ -176,7 +176,7 @@ remember({
 
 **If you complete a task without remembering anything, you likely missed something.** Review what you learned.
 
-For outdated KB entries → \`update(path, content, reason)\`
+For outdated AI Kit entries → \`update(path, content, reason)\`
 
 ---
 
@@ -287,7 +287,7 @@ Follow the **MANDATORY FIRST ACTION** and **Information Lookup Order** from code
 
 ## Research Methodology
 
-### Phase 1: KB Recall (BLOCKING)
+### Phase 1: AI Kit Recall (BLOCKING)
 \`\`\`
 search("task keywords")
 scope_map("what you need to investigate")
@@ -358,8 +358,8 @@ When invoked for a decision analysis, you receive a specific question. You MUST:
 
 ## Invocation Mode Detection
 
-- **Direct** (has KB tools) → Follow the **Information Lookup Order** from code-agent-base
-- **Sub-agent** (prompt has "## Prior KB Context") → Skip KB Recall, use provided context
+- **Direct** (has AI Kit tools) → Follow the **Information Lookup Order** from code-agent-base
+- **Sub-agent** (prompt has "## Prior AI Kit Context") → Skip AI Kit Recall, use provided context
 
 ---
 
@@ -387,7 +387,7 @@ Follow the **MANDATORY FIRST ACTION** and **Information Lookup Order** from code
 
 ## Review Workflow
 
-1. **KB Recall** — \`search("conventions relevant-area")\` + \`list()\` for past review findings, patterns
+1. **AI Kit Recall** — \`search("conventions relevant-area")\` + \`list()\` for past review findings, patterns
 2. **Blast Radius** — \`blast_radius\` on changed files to understand impact
 3. **FORGE Classify** — \`forge_classify\` to determine review depth
 4. **Review** — Evaluate against all dimensions below
@@ -452,7 +452,7 @@ Follow the **MANDATORY FIRST ACTION** and **Information Lookup Order** from code
 
 ## Review Workflow
 
-1. **KB Recall** — \`search("architecture decisions boundaries")\` + \`list()\` for past ADRs, patterns
+1. **AI Kit Recall** — \`search("architecture decisions boundaries")\` + \`list()\` for past ADRs, patterns
 2. **Analyze** — \`analyze_structure\`, \`analyze_dependencies\`, \`blast_radius\`
 3. **Evaluate** — Check all dimensions below
 4. **Report** — Structured findings with verdict
@@ -529,7 +529,7 @@ Trigger the decision protocol when there is an **unresolved non-trivial technica
 
 > Follow the FORGE (Fact-Oriented Reasoning with Graduated Evidence) protocol for all code generation and modification tasks.
 
-## KB Tools for FORGE
+## AI Kit Tools for FORGE
 
 | Tool | Purpose | When |
 |------|---------|------|

package/scaffold/definitions/tools.mjs

@@ -5,7 +5,7 @@
  * `ide` — IDE-native tool permissions, keyed by role. Each IDE adapter
  *          maps these to its own format (e.g., Copilot tool identifiers).
  *
- * When KB adds/removes a tool, update `kb` here — all agents pick it up.
+ * When AI Kit adds/removes a tool, update `kb` here — all agents pick it up.
  *
  * NOTE: The copilot adapter currently uses `aikit/*` wildcard instead
  * of listing individual tools. This array is kept as a reference and for future

package/scaffold/flows/aikit-advanced/skills/execute/SKILL.md

@@ -1,124 +1,124 @@
----
-name: execute
-description: Implement all tasks from the task breakdown, dispatching agents in parallel where possible.
----
-
-# Execution
-
-## Purpose
-
-Execute all tasks from the breakdown, dispatching implementation agents in batches for maximum parallelism while maintaining correctness.
-
-## Inputs
-
-- `.spec/<slug>/tasks.md` — the atomic task list with dependencies and agent assignments
-
-## Process
-
-1. **Load tasks** — Read task graph, dependencies, and parallelism map
-2. **Execute by batch** — For each batch in the parallelism map:
-   a. Dispatch assigned agents in parallel (different files = safe parallelism)
-   b. Each agent receives: task scope, affected files, acceptance criteria, relevant code context via `compact()`/`digest()`
-   c. Wait for all agents in batch to complete
-   d. Run `check({})` + `test_run({})` after each batch
-   e. Fix any failures before proceeding to next batch
-3. **Track progress** — Update task checkboxes as each completes
-4. **Handle failures** — If an agent reports `BLOCKED` or `NEEDS_CONTEXT`:
-   - Max 2 retries per task with refined context
-   - If still blocked, escalate to user
-5. **Final validation** — After all batches: `check({})` + `test_run({})` must pass
-
-## Outputs
-
-Produce `.spec/<slug>/progress.md` with:
-
-```markdown
-# Execution Progress: <feature title>
-
-## Task Status
-- [x] T1.1: <description> — DONE
-- [x] T1.2: <description> — DONE
-- [x] T2.1: <description> — DONE
-- [ ] T2.2: <description> — IN PROGRESS
-...
-
-## Changes Made
-| File | Change | Task |
-|------|--------|------|
-| <file> | <description> | T1.1 |
-| ... | ... | ... |
-
-## Tests Added/Modified
-<list of test files and what they cover>
-
-## Validation
-- check: PASS/FAIL
-- test_run: PASS/FAIL (<N> passed, <M> failed)
-
-## Deviations
-<any changes from the task plan and why>
-
-## Blocked Items
-<items that needed user intervention, if any>
-```
-
-## Agents
-
-| Agent | Role |
-|-------|------|
-| **Orchestrator** | Coordinates batch execution, handles failures and retries |
-| **Implementer** | Primary code writer for backend, logic, and infrastructure tasks |
-| **Frontend** | UI/UX implementation for React components, styling, responsive design |
-| **Refactor** | Code cleanup, restructuring, and pattern alignment tasks |
-
-**Parallelism rules**:
-- Read-only agents: unlimited parallelism
-- File-modifying agents: parallel ONLY on completely different files
-- Max 4 concurrent file-modifying agents
-
-## Foundation Integration
-
-Load these skills BEFORE executing this step:
-
-| Skill | Purpose | When |
-|-------|---------|------|
-| `aikit` | Core MCP tools — search, analyze, remember, validate | Always (auto-loaded) |
-| `present` | Rich rendering for any structured output — assessments, reports, comparisons, reviews, status boards, tables, charts, and all artifact content | Use for ANY output that benefits from rich rendering, not just dashboards |
-| `multi-agents-development` | Dispatch templates, task decomposition, review pipeline patterns | Before dispatching any subagent |
-| `brainstorming` | Structured ideation for design/creative decisions | Before any design choice or new feature exploration |
-
-### Presentation Rules
-- Use `present` for **any output** that benefits from rich rendering — not limited to dashboards
-- Assessments, reports, comparisons, reviews, status boards → `present({ format: "html" })`
-- Tables, charts, progress tracking, code review findings → always present
-- Artifact content and summaries → present with structured layout
-- Only use plain text for brief confirmations and simple questions
-
-### Orchestrator Dispatch Protocol
-
-Follow the `multi-agents-development` skill patterns for dispatch:
-
-1. **Independence Check** before parallelizing:
-   - Same files? → sequential
-   - Shared mutable state? → sequential
-   - Execution-order dependent? → sequential
-   - Need shared new types? → define contract first, then parallel
-   - All clear? → **parallel dispatch**
-
-2. **Subagent Context Template** (each dispatch includes):
-   - **Scope**: exact files + boundary (do NOT touch)
-   - **Goal**: acceptance criteria, testable
-   - **Arch Context**: actual code snippets via `compact()`/`digest()`
-   - **Constraints**: patterns, conventions, anti-patterns
-   - **Self-Review**: checklist before declaring DONE
-
-3. **Status Protocol**: `DONE` | `DONE_WITH_CONCERNS` | `NEEDS_CONTEXT` | `BLOCKED`
-4. **Max 2 retries** per task — then escalate to user
-
-## Completion Criteria
-
-- [ ] All tasks marked completed
-- [ ] `check({})` passes
-- [ ] `test_run({})` passes
-- [ ] No blocked items remaining
-- [ ] `.spec/<slug>/progress.md` written
+---
+name: execute
+description: Implement all tasks from the task breakdown, dispatching agents in parallel where possible.
+---
+
+# Execution
+
+## Purpose
+
+Execute all tasks from the breakdown, dispatching implementation agents in batches for maximum parallelism while maintaining correctness.
+
+## Inputs
+
+- `.spec/<slug>/tasks.md` — the atomic task list with dependencies and agent assignments
+
+## Process
+
+1. **Load tasks** — Read task graph, dependencies, and parallelism map
+2. **Execute by batch** — For each batch in the parallelism map:
+   a. Dispatch assigned agents in parallel (different files = safe parallelism)
+   b. Each agent receives: task scope, affected files, acceptance criteria, relevant code context via `compact()`/`digest()`
+   c. Wait for all agents in batch to complete
+   d. Run `check({})` + `test_run({})` after each batch
+   e. Fix any failures before proceeding to next batch
+3. **Track progress** — Update task checkboxes as each completes
+4. **Handle failures** — If an agent reports `BLOCKED` or `NEEDS_CONTEXT`:
+   - Max 2 retries per task with refined context
+   - If still blocked, escalate to user
+5. **Final validation** — After all batches: `check({})` + `test_run({})` must pass
+
+## Outputs
+
+Produce `.spec/<slug>/progress.md` with:
+
+```markdown
+# Execution Progress: <feature title>
+
+## Task Status
+- [x] T1.1: <description> — DONE
+- [x] T1.2: <description> — DONE
+- [x] T2.1: <description> — DONE
+- [ ] T2.2: <description> — IN PROGRESS
+...
+
+## Changes Made
+| File | Change | Task |
+|------|--------|------|
+| <file> | <description> | T1.1 |
+| ... | ... | ... |
+
+## Tests Added/Modified
+<list of test files and what they cover>
+
+## Validation
+- check: PASS/FAIL
+- test_run: PASS/FAIL (<N> passed, <M> failed)
+
+## Deviations
+<any changes from the task plan and why>
+
+## Blocked Items
+<items that needed user intervention, if any>
+```
+
+## Agents
+
+| Agent | Role |
+|-------|------|
+| **Orchestrator** | Coordinates batch execution, handles failures and retries |
+| **Implementer** | Primary code writer for backend, logic, and infrastructure tasks |
+| **Frontend** | UI/UX implementation for React components, styling, responsive design |
+| **Refactor** | Code cleanup, restructuring, and pattern alignment tasks |
+
+**Parallelism rules**:
+- Read-only agents: unlimited parallelism
+- File-modifying agents: parallel ONLY on completely different files
+- Max 4 concurrent file-modifying agents
+
+## Foundation Integration
+
+Load these skills BEFORE executing this step:
+
+| Skill | Purpose | When |
+|-------|---------|------|
+| `aikit` | Core MCP tools — search, analyze, remember, validate | Always (auto-loaded) |
+| `present` | Rich rendering for any structured output — assessments, reports, comparisons, reviews, status boards, tables, charts, and all artifact content | Use for ANY output that benefits from rich rendering, not just dashboards |
+| `multi-agents-development` | Dispatch templates, task decomposition, review pipeline patterns | Before dispatching any subagent |
+| `brainstorming` | Structured ideation for design/creative decisions | Before any design choice or new feature exploration |
+
+### Presentation Rules
+- Use `present` for **any output** that benefits from rich rendering — not limited to dashboards
+- Assessments, reports, comparisons, reviews, status boards → `present({ format: "html" })`
+- Tables, charts, progress tracking, code review findings → always present
+- Artifact content and summaries → present with structured layout
+- Only use plain text for brief confirmations and simple questions
+
+### Orchestrator Dispatch Protocol
+
+Follow the `multi-agents-development` skill patterns for dispatch:
+
+1. **Independence Check** before parallelizing:
+   - Same files? → sequential
+   - Shared mutable state? → sequential
+   - Execution-order dependent? → sequential
+   - Need shared new types? → define contract first, then parallel
+   - All clear? → **parallel dispatch**
+
+2. **Subagent Context Template** (each dispatch includes):
+   - **Scope**: exact files + boundary (do NOT touch)
+   - **Goal**: acceptance criteria, testable
+   - **Arch Context**: actual code snippets via `compact()`/`digest()`
+   - **Constraints**: patterns, conventions, anti-patterns
+   - **Self-Review**: checklist before declaring DONE
+
+3. **Status Protocol**: `DONE` | `DONE_WITH_CONCERNS` | `NEEDS_CONTEXT` | `BLOCKED`
+4. **Max 2 retries** per task — then escalate to user
+
+## Completion Criteria
+
+- [ ] All tasks marked completed
+- [ ] `check({})` passes
+- [ ] `test_run({})` passes
+- [ ] No blocked items remaining
+- [ ] `.spec/<slug>/progress.md` written