Execute

Use this skill when the work is understood and the job is to build it.

This is the canonical execution workflow. autopilot, ralph, ultragoal, ultrapilot, pipeline, and swarm route here.

Goal

Take a task from agreed intent to working code, with evidence that it works.

Workflow

Confirm the task is clear enough to build. If it is not, plan first.
Break the work into independent units; run genuinely independent units in parallel.
Implement the smallest correct change per unit, reusing existing utilities and patterns.
Verify as you go, not only at the end.
Report what changed, what was verified, and what remains.

Scale

Match the machinery to the task:

Single unit — implement directly, verify, done.
Several independent units — delegate to executor agents in parallel.
Long-running or unbounded — keep a durable task list and continue until the list is empty.
Needs coordinated parallel workers — use team.

Do not spin up coordination for work that one focused pass would finish.

Rules

Prefer deletion over addition when behavior is preserved.
Do not add dependencies without an explicit request.
Keep diffs small and reversible.
Placeholder TODOs, test.skip, and stub tests are blockers, not progress.
Authoring and approval are separate passes — do not self-approve; hand off to review or verify.

Completion

Before claiming done:

No pending tasks
Tests pass, or failures are reported plainly
Verification evidence collected

Output

Files changed
What was implemented
Evidence it works
What is still open

Files1

1 files · 1.0 KB

Select a file to preview

Overall Score

88/100

Grade

A

Excellent

Safety

95

Quality

85

Clarity

90

Completeness

78

Summary

This skill is a meta-workflow guide that directs AI agents through a canonical execution process: confirming task clarity, breaking work into units, implementing changes, and verifying outcomes. It does not perform direct file operations, shell commands, or external network calls — it is a decision tree and responsibility framework that agents follow when executing approved tasks.

Detected Capabilities

task delegationparallel coordinationworkflow orchestrationoutcome verificationhandoff routing

Trigger Keywords

Phrases that MCP clients use to match this skill to user intent.

execute approved taskbuild verified codeimplement featureverify implementationhand off for review

Use Cases

Execute a clearly-defined feature from planning to verified working code
Coordinate multiple independent implementation units in parallel
Manage long-running task lists until completion
Delegate sub-tasks to executor agents while maintaining oversight
Hand off code to review or verification workflows after implementation

Quality Notes

Skill provides clear decision tree for scale matching (single unit → several independent → long-running → coordinated workers)
Rules section explicitly forbids self-approval and emphasizes reversibility, preventing unvalidated commits
Output requirements are well-defined (files changed, implementation details, verification evidence, open items)
Workflow prioritizes verification during implementation, not only at the end
Completion checklist prevents claiming success prematurely (no pending tasks, tests pass, evidence collected)
Scope is intentionally non-prescriptive about *how* to implement details — focuses on *when* and *whether* to execute rather than tool usage
References other canonical skills (autopilot, ralph, ultragoal, review, verify) showing it fits within a larger ecosystem
Clean, action-oriented language with minimal bureaucracy

Model: claude-haiku-4-5-20251001Analyzed: Aug 31, 2026

Reviews

Add this skill to your library to leave a review.

No reviews yet

Be the first to share your experience.

execute