Project Management DocumentationRoadmapPlanningFeatures

Feature Matrix

Build and incrementally maintain a feature matrix document that maps RFC feature areas, stories, and deliverables from git history and codebase exploration.

Projects accumulate features over time but nobody maintains the map. Git log shows what changed, GitHub issues show what was planned, but nothing shows the full picture. This skill reconstructs the feature timeline from commits, issues, and codebase structure — then produces a living document grouped by RFC codes that maps every deliverable to its origin and status. Works with or without GitHub issues, integrates with github-project-manager when project-config.json exists, and supports incremental updates via a state footer so it never re-scans the full history.

Added Apr 10, 2026

$npx skills add johnefemer/skillfish --skill feature-matrix

What This Skill Can Do

Concrete capabilities you get when you install this skill.

Generate a complete feature inventory from git history, codebase structure, and GitHub issues

Group features into chronological RFC codes with descriptive titles and one-liner explanations

Map deliverables to GitHub issues when project-config.json exists, skip gracefully when it doesn't

Incrementally update the matrix after new features ship without full regeneration

Resolve file name conflicts when docs/features.md already has unrelated content

Track state via an invisible HTML comment footer for incremental maintenance across sessions

Real Conversations

See how the Feature Matrix skill responds to actual requests. These are representative examples of what your AI agent will produce.

You

/build-feature-matrix

Agent

I'll build a feature matrix for this project. Let me collect data from four parallel streams.

Stream 1 — Git Timeline:

git log --format="%h %ad %s" --date=short --reverse

Found 129 commits from 2026-03-08 to 2026-04-10.

Stream 2 — Codebase Scan: Found 25 page routes, 40+ API endpoints, 8 database migrations, 30+ components.

Stream 3 — GitHub Issues: Found .github/project-config.json — pulling issues. 8 issues (4 closed, 4 open) spanning auth, docs, extension, and skills.

Stream 4 — Existing Docs: No existing feature matrix found. Generating fresh.

Clustering into RFCs...

Migration dates (Mar 8, Mar 13, Mar 31, Apr 1) establish 4 anchor points. Commit clustering around these dates produces 14 RFC groups.

Written to docs/features.md:

## RFC-101: Foundation & Skills Platform — 2026-03-08 — Done
> Mono-repo scaffold, Astro site, and skills catalog with 167 migrated skills

### F-101.1: Project Foundation
Astro site with dark theme, Cloudflare Pages deployment, and mono-repo structure.

| # | Deliverable | Issues | Status |
|---|------------|--------|--------|
| 1 | Mono-repo scaffold with Astro, Tailwind, Cloudflare Pages deploy pipeline | — | Done |
| 2 | Landing page with hero, features overview, and neon-cyan brand theme | — | Done |
...

Totals: 14 RFCs, 30 features, 100+ deliverables. State footer written for incremental updates.

Get Started

How to install and use this skill in your preferred environment.

Skills are designed for AI coding agents (Claude Code, Cursor, Windsurf) and IDE-based workflows where the agent can read files, run scripts, and act on your codebase.

Models & Context

Which AI models and context windows work best with this skill.

Recommended Models

Works best with Claude Opus/Sonnet 4 or GPT-4o — models that can run parallel tool calls and synthesize large git histories. Smaller models work for incremental updates but may miss clustering patterns on full generation of 100+ commit repos.

Context Window

SKILL.md is ~12KB. For full generation, the agent reads git log output (1-5KB), codebase scan results (2-10KB), and optionally issues JSON (1-20KB). 100K+ context recommended for projects with 200+ commits.

Pro tips for best results

1

Be specific

Include numbers — users, budget, RPS — so the skill can size the architecture.

2

Share constraints

Compliance needs, team size, and existing stack all improve the output.

3

Iterate

Start with a high-level design, then ask follow-ups for IaC, cost analysis, or security review.

4

Combine skills

Pair with companion skills below for end-to-end coverage.

Works Great With

These skills complement Feature Matrix for end-to-end coverage. Install them together for better results.

$ skillfish add johnefemer/skillfish --all # install all skills at once

Ready to try Feature Matrix?

Install the skill and start getting expert-level guidance in your workflow — any agent, any IDE.

$npx skills add johnefemer/skillfish --skill feature-matrix
← Browse all skills