vc-plan-discovery

Discovers and groups plan artifacts by scanning frontmatter across feature folders.

2|Updated Sep 15, 2025
One-click install
npx skills add https://github.com/marsley01/Marsley-Portfolio-Web --skill vc-plan-discovery-marsley01
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: vc-plan-discovery
Source: https://github.com/marsley01/Marsley-Portfolio-Web/tree/main/.agents/skills/vc-plan-discovery
Command: npx skills add https://github.com/marsley01/Marsley-Portfolio-Web --skill vc-plan-discovery-marsley01

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Agents starting work on a feature often lack context about what plans already exist, what was tried before, and what is deferred. This Skill scans plan directories and surfaces all relevant plan artifacts grouped by status, so agents begin each task with full plan history instead of rediscovering it manually. ## Core Features & Use Cases - Scoped Plan Scanning: Reads all subfolders of the current feature folder, active-only plans from other features, and general-plans, following explicit scope rules. - Frontmatter-Only Extraction: Parses only the leading YAML frontmatter of each Markdown file, avoiding loading large files into context. - Grouped Output: Returns plans bucketed into Active Plans, Backlog, Completed, Reports, and References, with a summary count line and optional JSON output. - Use Case: At the start of a research or execution loop step, run the discovery script with a feature name to instantly see every active, deferred, and completed plan plus related reports and references for that feature. ## Quick Start Run the plan discovery script for my current feature and list all active, backlog, and completed plans with their descriptions.

Frequently Asked Questions about vc-plan-discovery

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

FAQPage Schema
How do I find all existing plans for a feature before starting work?

Run the discover-plans.mjs script with the --feature flag pointing at your feature folder. It scans the same-feature folder at full depth, other features' active plans, and general-plans, then prints grouped results with names and descriptions.

How to get machine-readable plan discovery output?

Pass the --json flag to the discover-plans.mjs script. It returns a JSON object keyed by bucket (active, backlog, completed, reports, references) with each entry's path and parsed frontmatter fields.

Does the plan discovery script read entire Markdown files?

No, it extracts only the leading YAML frontmatter block of each .md file, reading at most the first 80 lines. This keeps context usage small and avoids loading large plan documents.

What happens if a plan folder or file has no frontmatter?

Files without valid frontmatter are skipped and logged as having no frontmatter. Missing root directories never cause errors; the script exits 0 unless given an invalid flag.

When should plan discovery run in an agent workflow?

It should run alongside context discovery at the start of every loop step such as research, validate, execute, or update-process, and any time an agent needs to know what plans exist or what was tried before.