โ† AI library

Editorial recipe โ€” not benchmarked โ€” reviewed September 1, 2026

Build an evidence-linked competitor scan

For founders and product teams making a bounded market or positioning decision.

Stack
Claude Code + Claude Opus 5 โ†’ Playwright MCP (isolated) โ†’ Obsidian โ†’ OpenAI Codex

Run this recipe

Work through it here

Progress stays in this browser. The downloaded Markdown kit works in any notes app or repository.

0/5 complete

Procedure

  1. Write one decision question, the comparison fields, named competitors, date cutoff, and an allowlist of primary-source domains before browsing.
  2. Use Playwright MCP in an isolated, logged-out browser profile to collect public vendor pages and original repositories; record the URL, access date, evidence type, and a short supporting note for each claim.
  3. Ask Claude Code to produce claim-ledger.md and brief.md, keeping facts, vendor claims, inferences, conflicts, and missing evidence visibly separate.
  4. Give Codex the ledger, brief, and captured sources; request a review of uncited claims, inconsistent comparisons, stale evidence, and conclusions that exceed the source material.
  5. Reopen every source supporting a decision-changing claim, resolve the review findings, and publish the brief with an as-of date and explicit unknowns.

Acceptance artifact

A dated competitor brief, primary-source claim ledger, conflict log, and resolved second-model review

Do not use it blindly

Web pages are untrusted input and Playwright MCP is not a security boundary; do not use authenticated sessions, obey access restrictions, and never let page content authorize downloads, commands, purchases, or data disclosure.

Evidence and setup