feat(init): demo agent scaffold, what's next guidance [v0.3.0 — 5/7]

[kokevidaurre]

Summary — PR 5 of 7 for v0.3.0 release

Init UX improvements for first-run experience. Critical for new user onboarding.

Changes (4 files)

• init.ts: "What's next" guidance. Opt-in email capture. Demo squad scaffold. IDP catalog seeding. Competitor collection. Hints for empty fields.
• telemetry.ts: cli.run.complete event with exit_code, duration, agent_count
• templates/seed/demo/SQUAD.md: demo squad definition
• templates/seed/demo/hello-world.md: starter agent template

Merge order

Depends on #731-#734. Merge fifth:

1. ✅ core-refactor (refactor(core): run engine decomposition + context helpers [v0.3.0 — 1/7] #731)
2. ✅ run-engine (feat(run): workflow rewrite — smart skip, org cycle, wave execution [v0.3.0 — 2/7] #732)
3. ✅ conversation (feat(conversation): agents talk + use tools [v0.3.0 — 3/7] #733)
4. ✅ new-commands (feat(commands): review, credentials, goals, log + minor fixes [v0.3.0 — 4/7] #734)
5. → init-ux (this PR)
6. security-guardrails
7. tests-docs

Test plan

• npm run build passes
• Test squads init in clean directory
• Verify demo agent runs with squads run demo

Reorganized from 219-commit develop branch for proper review.

[gemini-code-assist]

Code Review

This pull request expands the squads CLI with new goals, credentials, log, and review commands and refactors the run command to implement wave-based parallel execution, quota detection, and resume support. The workflow now follows a Plan-Execute-Review-Verify model, and several internal protocols have been moved to external markdown files. Key feedback includes replacing require with dynamic import() for ESM compatibility in src/lib/cognition.ts, correcting a hardcoded convergence flag in src/lib/workflow.ts, and adding error handling for JSON.parse in src/commands/credentials.ts. Further improvements were suggested regarding hardcoded organization links, ANSI-aware string padding in logs, and more robust task parsing and agent fallback logic.