- Home
- Skills
- Workflow Automation
- flightplan
flightplan
Plan, route, and sequence complex development work into checkable, tool-optimized tickets.
$5
Works with the AI Agents you already use
flightplan
Example session with this skill installed
/flightplan I need to migrate our auth system from local strategy to Auth0, including the database schema changes and frontend login components.
- Read your context and instructions
- Compiled the flightplan
- Generated the UI component
I have analyzed the project structure and identified the auth migration requirements. I've created a plan with 8 tickets, routing high-risk database changes to Opus and UI components to Sonnet.
Flightplan: Auth0 Migration
The plan is live. You can track progress and copy prompts from the plan page.
| Ticket | Task | Capability | Model |
|---|---|---|---|
| T-01 | Define Auth0 Tenant Config & Scopes | none | haiku |
| T-02 | Add @auth0/auth0-react dependency | none | haiku |
| T-03 | Migrate User schema to include auth0_id | postgres-mcp | opus |
| T-04 | Implement Auth0Provider in App.tsx | web-design | sonnet |
| T-05 | Create Login/Logout buttons | web-design | sonnet |
| T-06 | Update API middleware to verify JWTs | api-skill | opus |
| T-07 | Refactor Profile component for Auth0 data | none | sonnet |
| T-08 | End-to-end auth flow verification | testing | sonnet |
Next steps
- Open the plan page:
~/.claude/flightplan/jobs/github.com/acme/app/plan.html - Copy the prompt for T-01 from the page.
- Run
/model haikuthen paste the prompt to begin.
App.tsx
TSX · React component
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Development speed suffers when AI agents try to do everything at once, leading to context drift and messy code. This skill prevents that by separating planning from execution, breaking down complex tasks into a sequence of verified, checkable tickets.
What it does
- Multi-step planning generates a dependency-ordered sequence of 5 to 15 checkable tickets for complex tasks.
- Intelligent routing assigns each ticket to the specific skill, subagent, or MCP tool best suited for that specific sub-task.
- Model optimization selects the appropriate model and effort level based on the task's complexity and risk.
- Progress tracking maintains a live HTML plan page to log outcomes, evidence, and status for every ticket.
- Outcome verification judges finished work against strict "Done" criteria before moving to the next ticket.
How it works
- The /flightplan command triggers a deep read of the project context, git logs, and documentation to establish scope and success criteria.
- The "Grill" phase tests assumptions with up to four targeted questions to close gaps in the requirements.
- The Serve phase generates a precise prompt for the current ticket, which the user pastes to execute the work.
- The Record phase validates the result and updates the plan page, either proceeding to the next ticket or flagging a failure.
Frameworks & tools
Built for Claude Code. Integrates with the Skill, Agent, AskUserQuestion, and Artifact tools. Manages local state via ~/.claude/flightplan/.
Why this beats prompting it yourself
Manual prompting for long tasks often leads to "lazy" code or missed edge cases as the context window fills up. This skill enforces a pause between tickets for user verification and allows for switching models between planning and execution to save costs.
Use cases
- Feature kick-offs to break a large PR into smaller, reviewable commits.
- Bug hunting to sequence investigation, reproduction, and fixing steps.
- System refactoring to manage risky, multi-file changes without breaking functionality.
- Resume work to pick up exactly where a previous session left off with full context.
Known limitations
Does not perform the actual coding or implementation work itself. Requires a git remote or local path to generate project keys. Plan pages require the Artifact tool for the best experience.
How to use
Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scanned
Verified clean 7 days ago
- Passed all security checks, Safe to install
Requires