feat(plans): add MCP/skills integration phases and pause/resume to build workflow

Build command now includes:
- Phase 3.5: Skills connection with pause/resume for custom creation
- Phase 4.5: MCP server integration (connect existing, guide creation)
- Resume mechanism via build-state.json for pausing mid-build
- Explicit deployment target selection with trade-offs

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

.claude/plans/execution-guide.md

@@ -296,7 +296,16 @@ Read these files FIRST:
 
 Execute steps 8, 26, 27:
 8.  Update build command for domain templates and new features (Phase 0 template selection)
-26. Update build command to integrate ALL Phase 2-5 features (memory, proactive, governance, goals, org-chart, budget, heartbeat, Docker)
+26. Update build command to integrate ALL Phase 2-5 features:
+    - Phase 2.5: Memory setup
+    - Phase 3.5: Skills & custom components (with pause/resume via build-state.json)
+    - Phase 3.7: Proactive agent (ADL/VFM)
+    - Phase 4.5: Integrations & MCP servers (connect existing, guide creation, pause/resume)
+    - Phase 5.5: Governance (autonomy levels, approval gates, budget)
+    - Phase 5.7: Goals and org chart
+    - Phase 6: Deployment target selection (ALL targets with trade-offs, user MUST choose)
+    - Phase 7: Feedback loop setup
+    - Resume mechanism: --resume flag reads build-state.json
 27. Update plugin.json to v1.0.0, rewrite CLAUDE.md and README.md for full Agent Factory
 
 Steps 8 and 26 both modify commands/build.md — execute them SEQUENTIALLY.

.claude/plans/ultraplan-2026-04-11-agent-factory.md

@@ -906,12 +906,25 @@ graph TD
 - **Files:** `commands/build.md`
 - **Changes:** Extend the build workflow with new optional phases (user can skip any):
   - **Phase 2.5 (after Operating Manual, before Agent Team): Memory Setup** -- Ask if user wants 3-tier memory. If yes, generate SESSION-STATE.md, MEMORY.md, and memory/ directory from templates. Configure daily log rotation.
-  - **Phase 3.5 (after Agent Team): Proactive Agent** -- Ask if any agents should be proactive (self-improving). If yes, add ADL/VFM rules from proactive templates. Show VFM scoring example.
+  - **Phase 3.5 (after Agent Team): Skills & Custom Components** -- Ask if agents need specialized skills beyond what was generated. For each: check if a matching skill exists (Glob for `.claude/skills/`). If yes, wire it up. If no, offer two options: (a) generate a skill skeleton from template, (b) explain how to create one manually and offer to pause. If user says "I need to build this first", save build state to `build-state.json` (current phase, completed phases, all user choices so far) and say "Run `/agent-factory:build --resume` when ready." On resume: read build-state.json and continue from where we left off.
+  - **Phase 3.7 (after Skills): Proactive Agent** -- Ask if any agents should be proactive (self-improving). If yes, add ADL/VFM rules from proactive templates. Show VFM scoring example.
+  - **Phase 4.5 (after Pipeline): Integrations & MCP Servers** -- Critical phase for connecting external services. Ask: "What external services do your agents need? (Slack, GitHub, databases, APIs, etc.)". For each service:
+    1. Check if MCP server exists: search known servers (`@anthropic-ai/mcp-server-*`, community servers)
+    2. If exists: generate `.mcp.json` entry with correct config, explain what env vars are needed
+    3. If not exists: explain what an MCP server is, show the 3 options: (a) use an existing community server (provide search tips), (b) create a custom one (show skeleton + link to `/mcp-builder` skill or MCP docs), (c) use Bash tool directly (simpler but less structured)
+    4. If user needs to create an MCP server: save build state and pause (same resume mechanism as Phase 3.5)
+    5. After all integrations configured: validate `.mcp.json` syntax, test connectivity where possible
   - **Phase 5.5 (after Security): Governance** -- Ask autonomy level (0-4). Generate GOVERNANCE.md with approval gates matching the chosen level. Integrate budget tracking if user wants it.
   - **Phase 5.7: Goals and Org Chart** -- For multi-agent systems (3+ agents): generate GOALS.md and ORG-CHART.md. Define reporting structure and goal hierarchy.
-  - **Phase 6 update**: Add Docker as deployment option. Add `/schedule` with heartbeat as the primary recommended option. Reference deployment-advisor agent.
+  - **Phase 6 update: Deployment Target Selection** -- Present ALL deployment options with clear trade-offs:
+    - `/schedule` (Claude Code native) -- simplest, no infra needed, but requires Claude Code running
+    - Local cron/launchd -- runs without Claude Code open, but tied to one machine
+    - VPS systemd -- always-on, remote, but needs server access
+    - Docker -- portable, isolated, reproducible, but needs Docker installed
+    User MUST choose at least one. For each chosen target: generate config from templates, provide exact activation commands, explain how to verify it's running.
   - **Phase 7 update**: After test run, offer to set up feedback loop (performance scoring template). Show how to run pipeline-optimizer.sh after first week.
-  - **Summary update**: Include all new components in the final summary (memory, governance, goals, org-chart, budget, heartbeat).
+  - **Resume mechanism**: `build-state.json` format: `{ "phase": "4.5", "completed": ["1","2","3","3.5","4"], "choices": { "template": "monitoring", "agents": [...], "memory": true, ... }, "paused_reason": "User creating custom MCP server" }`. On `--resume`: read state, print summary of what's done, continue from saved phase.
+  - **Summary update**: Include all new components in the final summary (memory, governance, goals, org-chart, budget, heartbeat, MCP integrations, deployment target).
 - **Reuses:** Existing Phase 1-7 structure; all templates from Steps 9-18
 - **Verify:** `wc -l /Users/ktg/repos/agent-builder/commands/build.md` → expected: significantly more lines than original (was ~390, should be ~600+)
 - **On failure:** revert -- build command is critical path, must be valid