| name | exst - Execution Stage |
| description | Turns decisions into action. Breaks down tasks, sequences them, sets milestones, identifies the first action, and biases toward starting over perfecting the plan. |
| output | {"format":"prose"} |
Execution Stage
Input: $ARGUMENTS
Step 1: Break the Decision Into Tasks
Decompose the committed decision into concrete, doable tasks.
DECISION: [what was decided — one line]
TASK BREAKDOWN:
1. [Task 1]: [specific action] — estimated effort: [small/medium/large]
2. [Task 2]: [specific action] — estimated effort: [small/medium/large]
3. [Task 3]: [specific action] — estimated effort: [small/medium/large]
4. [Task 4]: [specific action] — estimated effort: [small/medium/large]
5. [Task 5]: [specific action] — estimated effort: [small/medium/large]
Rules:
- Each task should be completable in one sitting
- If a task feels too big, split it further
- Use verbs: "write," "build," "contact," "test" — not "think about" or "consider"
Step 2: Sequence the Tasks
Order tasks by dependencies, not by preference.
SEQUENCE:
Phase 1 — Foundation:
[ ] [task] — blocks: [what depends on this]
[ ] [task] — blocks: [what depends on this]
Phase 2 — Build:
[ ] [task] — requires: [what must be done first]
[ ] [task] — requires: [what must be done first]
Phase 3 — Finish:
[ ] [task] — requires: [what must be done first]
PARALLEL WORK: [tasks that can happen simultaneously]
Rule: If no task has dependencies, the sequence doesn't matter — start with the easiest one for momentum.
Step 3: Assign Resources
Identify what each task needs to get done.
RESOURCE MAP:
| Task | Who | Tools/Materials | Time | Blockers |
|------|-----|----------------|------|----------|
| [task 1] | [person/role] | [what's needed] | [estimate] | [any blockers] |
| [task 2] | [person/role] | [what's needed] | [estimate] | [any blockers] |
| [task 3] | [person/role] | [what's needed] | [estimate] | [any blockers] |
SKIP: If this is a solo project with no resource constraints, skip.
Step 4: Set Milestones
Define checkpoints to track progress and catch problems early.
MILESTONES:
| Milestone | Target Date | Success Criteria | Risk |
|-----------|-------------|-----------------|------|
| [milestone 1] | [date/timeframe] | [how to know it's done] | [what could go wrong] |
| [milestone 2] | [date/timeframe] | [how to know it's done] | [what could go wrong] |
| [milestone 3] | [date/timeframe] | [how to know it's done] | [what could go wrong] |
FINAL DELIVERABLE: [what "done" looks like for the whole effort]
Rule: Each milestone should be verifiable — you can point to evidence that it's complete.
Step 5: Identify the First Action
The most important output of an execution plan is the very first thing to do.
FIRST ACTION:
- What: [the single next physical action]
- When: [right now / today / by tomorrow]
- How long: [estimated time to complete]
- Why this first: [why this is the right starting point]
- Done when: [specific completion criteria]
Rules:
- The first action should be small enough to start immediately
- It should be concrete enough that there's no ambiguity about what to do
- It should create momentum — once started, the next action becomes obvious
Step 6: Begin
Bias toward action. The plan is good enough.
EXECUTION READINESS:
- Plan complete enough to start: [yes — if no, what's missing?]
- First action clear: [yes/no]
- Biggest risk: [what's most likely to go wrong]
- Mitigation: [how to handle the biggest risk]
- Perfectionism check: Are you delaying because the plan isn't perfect? [yes/no]
→ If yes: Start anyway. Adjust as you go.
STATUS: READY TO EXECUTE
FIRST ACTION: [repeat the first action here]
Integration
Use with:
/dcst -> Make the decision before planning execution/rfst -> Reflect after execution is complete/to -> For more detailed task organization/de -> For more comprehensive project decomposition
