Claude Code / Codex CLI skills I actually run in a PM work week — metric attribution, PRD review panel, cross-model GO/NO-GO gate, blind AB testing, audience-tailored briefs, decision triage, session handoff, Lark meeting notes, comment loops, peer-feedback drafting. Runnable engines, not prompt snippets.
ab-testing lark product-management ai-agents pm-tools feishu prompt-engineering prompt-testing ai-product-management llm-evals code-review-automation agent-skills peer-feedback claude-code agent-evaluation codex-cli claude-skills lark-cli prd-review metric-attribution
-
Updated
Aug 29, 2026 - JavaScript