# Executing Plans

> Use while carrying out an approved implementation plan, keeping task state, evidence, and deviations explicit.

- Skill: `dyai2025/executing-plans` (Agent Skill)
- Install (CLI): `npx skillmds@latest add dyai2025/executing-plans`
- Raw SKILL.md: https://api.skillmd.com/api/skills/dyai2025/executing-plans/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Productivity
- Author: DYAI2025 (https://skillmd.com/u/dyai2025)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/dyai2025/executing-plans

---


# Executing Plans

## Per-task loop
1. Read the current plan item and its acceptance signal.
2. Claim only one task at a time.
3. Execute using TDD where code changes are required.
4. Update evidence: commands, files changed, decisions, and blockers.
5. Mark complete only when the stated acceptance signal passes.

If reality diverges from the plan, stop and update the plan rather than silently improvising.