Workflow Lifecycle
How FlyDocs structures AI-assisted development into a repeatable lifecycle from capture to close.
FlyDocs structures AI-assisted development into a repeatable lifecycle. Every issue moves through defined stages with clear handoffs.
The stages
| Stage | Command | What happens |
|---|---|---|
| Capture | /capture | Create an issue (bug, feature, task) |
| Refine | /refine | Add acceptance criteria, priority, estimates |
| Activate | /activate | Assign and transition to implementing |
| Implement | /implement | Build against the spec with full context |
| Review | /review | Check implementation against acceptance criteria |
| Validate | /validate | QE verification (optional) |
| Close | /close | Mark done, record outcomes |
Sessions wrap it all together
/start-session and /wrap-session bookend your work:
- Start loads context, shows active issues, sets up the session
- Wrap summarizes progress, captures decisions, updates project state
This creates continuity. Every session builds on the last instead of starting from scratch.
Local and cloud
All workflow stages work with local file-based issues out of the box. No accounts, no API keys.
Coming soon: The cloud tier syncs with your existing project management tool (Linear, Jira, and more). Issues live in your tracker, FlyDocs handles the AI workflow layer on top. A web portal will give you project setup, team visibility, and session analytics across your whole team.
Join the Discord for upcoming features, support, and early access to what's next.
Next
- Skills: platform skills, community skills, and progressive disclosure
- Hooks & Guardrails: how deterministic enforcement works
- Context & Sessions: persistent context and session continuity