Claude Code / Codex CLI skills I actually run in a PM work week — metric change attribution, PRD review panel, cross-model GO/NO-GO review gate, blind AB testing for prompt changes, audience-tailored briefs, Lark meeting notes, comment loops and peer-feedback drafting. Real runnable engines, not prompt snippets.
ab-testing lark product-management ai-agents pm-tools feishu prompt-engineering prompt-testing anthropic ai-product-management llm-evals code-review-automation agent-skills peer-feedback claude-code agent-evaluation codex-cli lark-cli prd-review metric-attribution
-
Updated
Aug 5, 2026 - JavaScript