---
name: closed-loop-delivery
description: ALWAYS use this when a coding task needs to reach a real done state with acceptance criteria, implementation, verification, review follow-through, and runtime proof where applicable.
Closed-Loop Delivery
Selective Reading Rule
Start with:
references/usage-routing.md
references/quality-checklist.md
Then load only the inherited docs, scripts, assets, or examples that match the user's actual task.
Selective Reading Rule
Start with:
references/senior-master-standard.md
Start with references/definition-of-done-template.md.
Use it to define what counts as done before implementation gets too far ahead of verification.
Purpose
Drive a task to a proven outcome instead of stopping at code edits.
Workflow
Define acceptance criteria.
Isolate the work when risk warrants it.
Implement the smallest change that can satisfy the criteria.
Verify locally.
Resolve review feedback or runtime issues.
Repeat until the criteria are proven or the blocker is explicit.
Rules
No vague completion claims.
No skipping runtime proof when runtime behavior matters.
No hiding unresolved blockers in a success summary.
1---2name: closed-loop-delivery3description: ---4---5---6name: closed-loop-delivery7description: ALWAYS use this when a coding task needs to reach a real done state with acceptance criteria, implementation, verification, review follow-through, and runtime proof where applicable.8---910# Closed-Loop Delivery1112## Selective Reading Rule1314Start with:1516- `references/usage-routing.md`17- `references/quality-checklist.md`1819Then load only the inherited docs, scripts, assets, or examples that match the user's actual task.2021## Selective Reading Rule2223Start with:2425- `references/senior-master-standard.md`262728Start with `references/definition-of-done-template.md`.2930Use it to define what counts as done before implementation gets too far ahead of verification.3132## Purpose3334Drive a task to a proven outcome instead of stopping at code edits.3536## Workflow37381. Define acceptance criteria.392. Isolate the work when risk warrants it.403. Implement the smallest change that can satisfy the criteria.414. Verify locally.425. Resolve review feedback or runtime issues.436. Repeat until the criteria are proven or the blocker is explicit.4445## Rules4647- No vague completion claims.48- No skipping runtime proof when runtime behavior matters.49- No hiding unresolved blockers in a success summary.
Run npx skillmds@latest add anhvu1107/closed-loop-delivery in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.
--- It is listed under AI & ML on SkillMD.
This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.
This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.
Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.
anhvu1107 (@anhvu1107) published this skill. Their other Agent Skills are listed on their SkillMD profile.