Value Stories
How the unified CLI creates value. Each story is an intent flow: a scenario triggers an intention, actions produce artifacts, outcomes prove value.
Action — Human-initiatedCron — ScheduledHook — Event-triggered
Can agents use the CLI?
Runtime introspection and safe mutation preview. Agents compose workflows dynamically.
Does migration preserve behavior?
40+ commands must produce identical output after decomposition. Zero-disruption migration.
Are boundaries enforced?
Direct DB imports blocked at lint time. Convenience can't bypass the repository layer.
Kill Signal
No new CLI added in 90 days, or agents stop using plan-cli. Neither has triggered.
Who this is for
| Who | Job |
|---|---|
| Agent orchestrator | Discover + compose CLI commands at runtime without hardcoded paths |
| Operator | One binary — stop remembering 5 entry points |
| Contributor | ESLint blocks direct DB access — can't accidentally bypass repositories |
Questions
Can you decompose a monolith without a comprehensive integration test suite — and if not, what's the minimum suite that gives you confidence?
- What's the migration path for external consumers who import plan-cli directly?
- If the ESLint rule blocks direct DB access, what's the escape hatch for genuine emergencies?
- If every CLI shares the same platform lib, which current bugs disappear without being fixed?