| name | Executing Plans |
| description | Use to execute an implementation plan yourself, inline in this conversation, with review checkpoints. The simpler alternative to subagent-driven-development - use it for small plans or when the agents feature is off. Works in an isolated git worktree. |
Executing Plans
Load the plan, review it critically, execute every task in order, and report when complete. This is the inline path — you do the work yourself rather than dispatching sub-agents. For larger plans, or to keep context lean and run cheaper, prefer the subagent-driven-development skill instead.
Announce at start: "I'm using the executing-plans skill to implement this plan."
Step 1: Set up and review
- Isolated workspace. Invoke the using-git-worktrees skill (
open_skill → using-git-worktrees) to create and cd into .worktrees/<branch>. Never start on main/master without explicit user consent.
- Read the plan and review it critically. Note the Global Constraints. If you spot conflicts, gaps, or tasks that contradict each other or the constraints, raise them with the user before starting. If it's clean, proceed.
Step 2: Execute tasks
The plan is your checklist — its steps use - [ ] checkboxes, and the plan file on disk is your durable source of truth (it survives a context compaction; your memory does not).
For each task, in order:
- Work the steps exactly as written — the plan has bite-sized steps for a reason. Write the failing test, run it, write the minimal code, run the tests, commit.
- Run the verifications the task specifies — prefer an
msbuild__build_* tool for a .NET solution, otherwise the project's configured command via command__run. Don't skip them.
- Mark progress on disk: as each step completes, flip its
- [ ] to - [x] in the plan file with files__edit. After a compaction or a fresh start, re-read the plan — checked steps are done; resume at the first unchecked one. Don't re-do completed, already-committed work.
Step 3: Complete
After all tasks are done and verified, invoke the finishing-a-development-branch skill (open_skill → finishing-a-development-branch).
When to Stop and Ask
Stop immediately and ask the user — don't guess — when you:
- hit a blocker (missing dependency, a test that won't pass, an unclear instruction),
- find the plan has a critical gap that prevents starting or continuing,
- don't understand a step, or a verification fails repeatedly.
If the user updates the plan in response, return to Step 1's critical review before continuing.
Remember
- Review the plan critically first.
- Follow the steps exactly; don't skip verifications.
- Track progress in the plan's checkboxes, on disk.
- Stop when blocked — don't force through.
- Never start on
main/master without explicit consent.