From softpowers
Use when you have a written implementation plan to execute in a separate session with review checkpoints
How this skill is triggered — by the user, by Claude, or both
Slash command
/softpowers:executing-plansThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Load plan, review critically, execute phases and their sub-tasks, report when complete. Review gates belong at the end of each phase, not after every task/sub-task.
Load plan, review critically, execute phases and their sub-tasks, report when complete. Review gates belong at the end of each phase, not after every task/sub-task.
This is a delegated implementation mode. Use it only when the human explicitly chooses to have the agent implement the plan rather than using softassist.
Announce at start: "I'm using the executing-plans skill as delegated implementation for this plan."
Note: This is the inline fallback for delegated implementation. If the human chose delegated implementation, first inspect the session's subagent and workflow tools. If subagent dispatch is available, prefer softpowers:subagent-driven-development, where the controller builds an internal dependency schedule from the plan without rewriting it and uses parallel waves when the harness supports them. If subagents are unavailable, use this skill. If the human wants to stay the primary implementer, use softassist instead.
main and master as the default-branch candidates.New worktree or New branch hereContinue here, New worktree, or New branch hereNew worktree: use softpowers:using-git-worktreesNew branch here:
Continue here: verify the current branch is not the default branch, then proceed in placeFor each phase:
After all phases complete and verified:
STOP executing immediately when:
Ask for clarification rather than guessing.
Return to Review (Step 2) when:
Don't force through blockers - stop and ask.
Required workflow skills:
New worktreenpx claudepluginhub bnema/softpowersCreates structured, bite-sized implementation plans from specs or requirements before writing code. Useful for breaking down multi-step tasks into testable steps with file structure and task boundaries.