What problem does it solve?
Manual market research across Reddit, HN, X, GitHub, and other community sources is slow and unstructured, and agent-generated summaries often invent evidence. This Skill runs a deterministic research engine over real community posts and produces a single evidence-cited brief with strategic frameworks.
Core Features & Use Cases
- Community Signal Research: Delegates to the last30days engine via
oma market run to gather real posts from Reddit, HN, X, Bluesky, GitHub, Polymarket, arXiv, and more, with keyless sources always running.
- Intent-Based Frameworks: Classifies requests into pain, trend, competitor, or discovery intents and appends SWOT, Porter's Five Forces, or PESTEL sections citing only engine evidence clusters.
- Guardrails and Output Laws: Runs a detect-trap preflight that refuses invalid topics, enforces citation and formatting laws, and writes one dated brief to
.agents/results/market/.
- Use Case: Ask "What are the top VS Code pain points this month?" and receive a markdown brief with ranked evidence clusters, real upvoted quotes, and a SWOT analysis.
Quick Start
Ask the agent to research the pain points and community sentiment for a product or category, for example: run market research on VS Code user complaints over the last 30 days.