Skip to main content
Category: GTM Intelligence · Version: 0.0.1 · Author: opendirectory · Compatibility: All targets (claude, codex, opencode, gemini, anti-gravity, openclaw, hermes)
Give the skill a competitor’s public GitHub repo URL. It fetches their open issues, filters noise locally, clusters into 6 demand categories using the AI already running the skill, scores by real engagement (reactions), detects ignored demand (high reactions + no response = your opportunity), and outputs a ranked demand gap report with a GTM messaging brief.

​
Install

​
What It Does

  • Fetches up to 200 open issues from any public GitHub repo (no auth required)
  • Filters noise locally: removes PRs, bot issues, zero-engagement issues, and chore-pattern titles
  • Computes a demand score per issue: (reactions["+1"] x 2) + (comments x 0.5)
  • Detects “ignored demand”: issues with 10+ reactions, open 6+ months, no planned label
  • Clusters issues into 6 categories using the AI running the skill (no external API key needed): feature gaps, bug patterns, UX complaints, performance issues, missing integrations, missing docs
  • Generates 5-8 cluster themes with total demand scores
  • Messaging brief: 3 positioning angles + 3 outreach hooks + cluster-specific headlines
  • Saves output to docs/demand-signals/[owner]-[repo]-[date].md

​
Requirements

No external AI API key needed. The skill uses the AI assistant running it (Claude, Gemini, Copilot) to do the clustering and messaging brief. GitHub token is optional but recommended for repeated use.

​
Setup

​
How to Use

​
Why This Matters

A busy product team cannot read 500 open issues. This skill does the work: ranks by real user demand (reaction counts), clusters into actionable GTM categories, detects ignored demand (where competitors have been silent the longest), and translates raw complaints into positioning language. The demand score formula: (reactions["+1"] x 2) + (comments x 0.5). Reactions are weighted higher than comments because reactions are pure signal — a thumbs-up means “I want this.” Comments include maintainer responses, off-topic discussion, and spam. Ignored demand is your opportunity: An issue with 50 reactions that has been open for 2 years with no planned label means their users have been waiting. That is your product’s first billboard.

​
The 6 Demand Categories

​
Output

Each run produces:
  1. Demand Gap Leaderboard: clusters ranked by total demand score
  2. Ignored Demand section: issues open 6+ months with 10+ reactions and no response
  3. Top 10 Highest-Demand Issues: verbatim titles, reaction counts, direct GitHub links
  4. Cluster Deep Dives: top 3 clusters with theme name, pain summary, top 3 verbatim issues
  5. Messaging Brief: 3 positioning angles citing exact demand evidence
  6. GTM Angles: 3 outreach hooks using verbatim issue language

​
Cost per Run

  • GitHub API: 2 calls — free (unauthenticated: 60/hr limit, token: 5000/hr)
  • AI analysis: uses the model already running the skill — no additional cost
  • Total: free

​
Project Structure

​
License

MIT