azclaude-copilot 0.1.4 → 0.1.6

package/README.md

@@ -424,7 +424,7 @@ azclaude-copilot/
 ├── DOCS.md                          <- full user guide
 ├── SECURITY.md                      <- security policy + architecture
 ├── tests/
-│   └── test-features.sh          ← 1041 tests
+│   └── test-features.sh          ← 1048 tests
 ```
 
 ---
@@ -450,11 +450,11 @@ The runner is stateless. These files ARE the state.
 
 ## Verified
 
-1041 tests. Every template, command, capability, agent, and CLI feature verified.
+1048 tests. Every template, command, capability, agent, and CLI feature verified.
 
 ```bash
 bash tests/test-features.sh
-# Results: 1041 passed, 0 failed, 1041 total
+# Results: 1048 passed, 0 failed, 1048 total
 ```
 
 ---

package/bin/cli.js

@@ -287,7 +287,7 @@ function installGlobalHooks(cli) {
 // ─── Capabilities ─────────────────────────────────────────────────────────────
 
 // Core capability dirs installed by default; advanced dirs only with --full
-const CORE_CAP_DIRS  = ['shared', 'level-builders'];
+const CORE_CAP_DIRS  = ['shared', 'level-builders', 'evolution'];
 const FULL_CAP_DIRS  = ['shared', 'level-builders', 'evolution', 'intelligence'];
 
 function installCapabilities(projectDir, cfg, full) {
@@ -317,7 +317,7 @@ function installCapabilities(projectDir, cfg, full) {
     copyDir(path.join(src, dir), path.join(dst, dir));
   }
   ok(`Capabilities installed (${cfg}/capabilities/) — ${full ? 'full' : 'core'}`);
-  if (!full) info('Run npx azclaude --full to add evolution + intelligence capabilities');
+  if (!full) info('Run npx azclaude --full to add intelligence capabilities (debate, pipeline, ELO)');
   info('manifest.md is your capability index — read it to find what to load');
 }
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "azclaude-copilot",
-  "version": "0.1.4",
+  "version": "0.1.6",
   "description": "Autonomous product builder. Describe once, AZCLAUDE builds it across sessions — planning, implementing, testing, evolving, deploying. Zero human input.",
   "bin": {
     "azclaude": "./bin/cli.js",

package/templates/commands/evolve.md

@@ -32,13 +32,15 @@ If $ARGUMENTS is blank or "full": run all cycles below.
 ls .claude/agents/loop-controller.md 2>/dev/null
 ```
 
-**If loop-controller.md exists**: delegate ALL evolution work to it via Agent tool:
+**If loop-controller.md exists**: delegate evolution work to it via Agent tool:
 ```
 Run a full evolution cycle: re-derivation check, then Cycle 1 (detect/generate/evaluate),
 Cycle 2 (knowledge consolidation if 3+ sessions), Cycle 3 (topology if friction detected).
 Show the full cycle report when done.
 ```
-**STOP HERE** — do not run the steps below.
+**After loop-controller finishes**: continue to Step 7 (Generate Project-Specific Skills and Agents).
+The loop-controller handles gap detection and code fixes, but skill/agent generation requires
+reading level-builders which the loop-controller does not do.
 
 ---
 
@@ -150,7 +152,37 @@ If the file doesn't exist, create it with this header:
 
 ---
 
-## Step 7: Promote GENERAL Skills
+## Step 7: Generate Project-Specific Skills and Agents from Evidence
+
+### Skills from git evidence
+```bash
+git log --name-only --format="" --diff-filter=AM | sort | uniq -c | sort -rn | head -20
+```
+If a file pattern appears 5+ times (e.g., `src/components/*.tsx`, `content/*.md`, `api/routes/*.py`):
+1. Read `.claude/capabilities/level-builders/level3-skills.md`
+2. Create a skill that encodes the workflow for that pattern
+3. Save to `.claude/commands/{skill-name}.md`
+4. Only create if no existing skill covers this workflow
+
+### Agents from co-change clusters
+```bash
+git log --name-only --format="" --diff-filter=M | sort | uniq -c | sort -rn | head -30
+```
+If 3+ files in the same directory change together across 3+ commits:
+1. Read `.claude/capabilities/level-builders/level5-agents.md`
+2. Create an agent with all 5 layers in `.claude/agents/cc-{name}.md`
+3. Use co-change data for scope boundaries
+4. Only create if no existing agent covers this cluster
+
+### Check existing before creating
+```bash
+ls .claude/commands/*.md .claude/agents/*.md 2>/dev/null
+```
+Skip creation if a skill/agent already covers the same workflow.
+
+---
+
+## Step 8: Promote GENERAL Skills
 
 For any fix tagged GENERAL in EVALUATE:
 1. Copy to `~/shared-skills/{name}.md`

package/templates/commands/setup.md

@@ -70,7 +70,49 @@ The orchestrator-init agent:
 
 ---
 
-## Step 5: Quality Gate
+## Step 5: Generate Project-Specific Skills
+
+Read `.claude/capabilities/level-builders/level3-skills.md` for the skill design guide.
+
+Analyze the project's domain, stack, and recurring workflows:
+1. Check what file types and patterns exist (endpoints, pages, components, schemas, tests)
+2. Check git log for repeated commit patterns (what work is done most often)
+3. Identify 2-4 project-specific skills that would encode the most common workflows
+
+For each skill:
+- Create `.claude/commands/{skill-name}.md` with pushy description (3+ trigger variants)
+- Follow RECIPE pattern (steps to do something, not documentation)
+- Include completion rule
+- ≤ 500 lines (references/ subdir for overflow)
+
+**Stack-specific defaults** (generate if stack detected):
+| Stack | Generate |
+|-------|---------|
+| Next.js / React | `new-page.md`, `new-component.md` |
+| Express / FastAPI | `new-endpoint.md` |
+| Any DB | `migrate.md` |
+| Any deploy config | `deploy.md` |
+| Markdown content | `new-content.md`, `validate-content.md` |
+| Exam / quiz | `new-exam.md`, `validate-exam.md` |
+
+**TaskUpdate → completed** for Create skills.
+
+---
+
+## Step 6: Generate Project-Specific Agents (if project has 10+ files)
+
+Read `.claude/capabilities/level-builders/level5-agents.md` for agent design guide.
+
+1. Run co-change analysis: `git log --name-only --format="" --diff-filter=M | sort | uniq -c | sort -rn | head -30`
+2. Identify file clusters that change independently → each cluster = potential agent
+3. Create 1-3 agents in `.claude/agents/cc-{name}.md` with all 5 layers
+4. Prefix all agents with `cc-` to avoid framework collisions
+
+Skip if project has < 10 files or < 5 commits.
+
+---
+
+## Step 7: Quality Gate
 
 Load `capabilities/shared/quality-check.md` and run the full environment check.