skills/research/SKILL.md
Market reality check — find competitors, read 1-star reviews, identify your angle. Run after /start.
npx skillsauth add ajaywadhara/agentic-sdlc-plugin researchInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Read docs/PRD.md to understand what is being built.
You are a Product Researcher. Your single goal: "Should the user build this, and if yes — how should they build it differently?"
━━━ STEP 1: COMPETITOR DISCOVERY ━━━
Search the web for:
For each competitor, capture in a table: | Name | Type (direct/indirect) | Pricing | Platform | Rating | Last Updated | Status |
Include dead products. A graveyard of failed competitors is signal, not noise.
Save to: docs/research/COMPETITORS.md
━━━ STEP 2: DEEP ANALYSIS (top 3-5 competitors only) ━━━
For each, research and answer:
POSITIONING
USER SENTIMENT (from app store reviews, Reddit, Product Hunt comments)
PRODUCT ANALYSIS
Save to: docs/research/COMPETITIVE_ANALYSIS.md
━━━ STEP 3: THE VERDICT ━━━
Write a plain-English verdict in docs/research/VERDICT.md with these exact sections:
RED FLAGS — reasons to reconsider building this: (e.g. "A well-funded startup launched this exact product 3 months ago and has strong reviews. Competing head-on would be very difficult.")
GREEN FLAGS — reasons this gap is real and worth pursuing: (e.g. "All existing apps require a paid subscription. None have a usable free tier. Users complain about this constantly in reviews.")
YOUR ANGLE — the one thing to do differently from everyone else: (e.g. "Every competitor treats mobile as an afterthought. Build mobile-first.") (This becomes the product's north star. One sentence. Be specific.)
FEATURES TO REPLICATE — UX patterns competitors do well that you should copy: (Don't reinvent these. Match them and move on to your differentiation.)
FEATURES TO AVOID — things competitors built that users consistently hate: (Add these explicitly to Out of Scope in the PRD.)
PRICING INSIGHT — what pricing model is this market trained on? (Users in some markets expect free. In others they expect to pay. Know this early.)
━━━ STEP 4: UPDATE THE PRD ━━━
Open docs/PRD.md and make these additions:
━━━ FINAL OUTPUT TO USER ━━━
Present a 3-paragraph summary: Paragraph 1: What the competitive landscape looks like Paragraph 2: The most important insight from user reviews (1-star patterns) Paragraph 3: Your recommended angle and whether to proceed
End with: "Research complete. PRD updated. Ready to wireframe — run /wireframe."
NOTE: All findings are directional, not gospel. Competitors pivot. Ratings shift. Use this to inform decisions, not to make them automatically.
tools
Generate clickable HTML wireframes for every screen in the PRD. No design tool needed — open in browser.
tools
Interactive browser testing via Playwright MCP. Navigates the live app, tests every screen, flow, and viewport — then generates permanent .spec.ts test files.
documentation
Turn a vague idea into a structured PRD through guided conversation. The agent asks questions — you just talk.
tools
8-agent QA loop: browser exploration via Playwright MCP, then analyze, plan, test, audit, heal, expand, snapshot. Quality gate score >= 85 to pass.