Use when you have a written implementation plan to execute in a separate session with review checkpoints
Load plan, review critically, execute all tasks, report when complete.
Announce at start: "I'm using the executing-plans skill to implement this plan."
Note: This skill works much better with access to subagents. The quality of its work will be significantly higher if run on a platform with subagent support (such as Claude Code or Codex). If subagents are available, use subagent-driven-development instead of this skill.
Before starting implementation, read references/best-practices.md (relative to this skill's directory). This file contains 24 battle-tested patterns extracted from real code review fixes in this codebase — things like input validation, cascade-delete safety, soft-delete indexes, resource cleanup, and transaction scoping. These patterns reflect recurring mistakes that have been caught and fixed in production code, so applying them proactively saves review cycles and prevents bugs.
For each task:
STOP executing immediately when:
Ask for clarification rather than guessing.
Return to Review (Step 1) when:
Don't force through blockers - stop and ask.
Required workflow skills:
Bundled references: