plannotator-setup-goal

Turns an idea into a reviewed goal package with facts and an execution plan.

Updated Aug 2, 2026
One-click install
npx skills add https://github.com/mikalv/pi-extensions --skill plannotator-setup-goal-mikalv
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: plannotator-setup-goal
Source: https://github.com/mikalv/pi-extensions/tree/main/external/plannotator/apps/skills/extra/plannotator-setup-goal
Command: npx skills add https://github.com/mikalv/pi-extensions --skill plannotator-setup-goal-mikalv

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Vague feature ideas often go straight to implementation without shared understanding, causing scope creep and rework. This Skill converts a raw idea or objective into a structured goal package at goals/<slug>/ containing an interview record, a verified fact sheet, and a gated execution plan. ## Core Features & Use Cases - Structured Interview Bundle: Builds a compact question bundle with recommended answers and options, then collects answers through the Plannotator browser UI via plannotator setup-goal interview. - Fact Sheet Review: Derives testable, verifiable facts from interview answers, captures automated-verification selections, and persists them to facts.md and facts.meta.json. - Gated Execution Plan: Explores the codebase to produce plan.md with ordered steps, verification commands, and risks, then gates it through plannotator annotate --gate until approved. - Use Case: A developer has a rough idea like "add team workspaces". The Skill rearticulates the goal, interviews the user on scope and success criteria, produces an accepted fact sheet, and outputs a plan ready to launch with /goal goals/<slug>/goal.md. ## Quick Start Ask the agent to use the plannotator-setup-goal skill to turn your feature idea into a goal package for /goal.

Frequently Asked Questions about plannotator-setup-goal

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I turn a feature idea into an execution plan?▼

Run the setup-goal workflow: it rearticulates your idea, interviews you through a compact question bundle in the Plannotator UI, derives a reviewed fact sheet, then explores the codebase to write an ordered plan.md gated by plannotator annotate.

What files does the goal setup workflow produce?▼

It creates a goals/<slug>/ directory containing interview.json and interview-result.json, facts-review.json and facts-result.json, facts.md with facts.meta.json, plan.md, and a final goal.md referencing the facts and plan.

Can I skip the interview if my goal is already clear?▼

The interview bundle is the default path, but obvious answers are inferred rather than asked. An optional grill-first pass exists for vague goals, and it is explicitly opt-in so well-scoped goals go straight to the bundle.

What happens if the Plannotator browser session is dismissed?▼

If the interview or facts session is dismissed, the workflow stops and reports that the setup was closed. To revise after a session, update the working JSON file and rerun the command rather than reconstructing from memory.

How are facts verified in the generated plan?▼

Each fact carries an automatedVerification flag selected during the facts review. Facts marked true are treated as requiring concrete automated checks in the plan, unless a blocker is documented.