[plan] Strengthen noop documentation and ensure all workflow prompts include explicit noop instructions

[github-actions]

Objective

Strengthen documentation and examples for the noop safe-output tool, making it explicit that noop must be called when the agent determines no action is needed.

Context

From Agent Persona Exploration discussion #18026:

Strengthen noop documentation — Make explicit that noop must be invoked when the agent determines no action is needed; this is the #1 runtime failure mode for safe-output workflows.

The ci-doctor analysis confirms that missing noop calls is the most common cause of safe-output failures.

Approach

1. Review current noop documentation in docs/ and identify gaps
2. Update workflow prompt templates/examples to always include explicit noop instructions, e.g.:
   • "If no action is needed, call noop with a message explaining why"
   • Add a decision tree or checklist pattern to workflow prompts
3. Add a warning/note in the safe-outputs documentation section
4. Review .github/workflows/*.md example workflows and ensure each one includes explicit noop guidance in its prompt
5. Consider adding a linting/validation check that warns when a workflow prompt doesn't mention noop

Files to Modify

• docs/ — Add/update noop guidance in safe-outputs section
• .github/workflows/*.md — Ensure all example workflows mention noop explicitly in their prompts

Acceptance Criteria

• Documentation explicitly states noop is required (not optional) when no action is taken
• All example workflows in .github/workflows/*.md include explicit noop instruction in their prompts
• Documentation includes a clear example showing when to call noop vs other safe-output tools
• Failure mode is documented: "what happens if you forget noop"

Generated by Plan Command for issue #discussion #18026

• expires on Feb 26, 2026, 3:15 AM UTC