From product-playbook-for-agentic-coding
Quality check for AI-generated output. Detects and removes hedge stacking, throat-clearing, motivational padding, restating, and unnecessary transitions. Use before presenting any substantive artifact to the user. Don't use for code comments or commit messages (those have their own conventions).
How this skill is triggered — by the user, by Claude, or both
Slash command
/product-playbook-for-agentic-coding:conciseness-checkThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Run this check before presenting any substantive artifact (PRD, tech plan, research synthesis, critique, handoff note) to the user.
Run this check before presenting any substantive artifact (PRD, tech plan, research synthesis, critique, handoff note) to the user.
Scan the artifact for these patterns. When found, apply the fix immediately.
| Pattern | Example | Fix |
|---|---|---|
| Hedge stacking | "it's worth noting that perhaps" | State directly or omit |
| Unnecessary transitions | "with that in mind", "building on the above" | Delete — context is implicit from document structure |
| Throat-clearing | "in order to effectively address" | Start with the verb |
| Motivational padding | "this powerful approach enables teams to" | State what it does, drop the adjectives |
| Restating | Same point in different words in consecutive sentences | Keep the stronger version, delete the other |
| Meta-commentary | "I wanted to reach out to...", "as mentioned above" | Delete |
| Excessive hedging | "it might be possible that this could potentially" | Pick one hedge or commit to the statement |
Multi-phase commands that produce a substantive artifact should run a self-review pass before presenting to the user.
Apply this when the output is:
Skip for:
Fix issues silently. The user sees clean output, not the self-check process.
Any document going to multiple reviewers should use a two-layer structure.
Use a clear separator between the two layers:
---
## Appendix
[Reference material below]
Apply this structure when the artifact:
Skip for:
This skill runs as an internal self-check, not a user-facing suggestion:
No user prompt needed — apply silently as part of output quality control.
Creates, edits, and verifies skills using a test-driven development approach with pressure scenarios and subagents.
npx claudepluginhub daviswhitehead/product-playbook-for-agentic-coding-plugin --plugin product-playbook-for-agentic-coding