메뉴
Implement from an active plan or ExecPlan while keeping it up to date. Use when an agent needs to implement the active plan, continue plan-driven work, resume the current plan, or carry out an existing ExecPlan.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Author CLI surfaces as agent-facing documentation, especially `--help` and actionable error messages. Use with any language or parser design when adding or changing a CLI interface, improving help text or validation errors, moving command reference out of AGENTS.md / CLAUDE.md / a README / a skill and into CLI surfaces, or making errors suggest the corrected command line or tool call.
Create or revise repository `AGENTS.md` files using a progressive-disclosure, dispatcher-style structure. Use when an agent needs to author or edit a top-level or directory-local `AGENTS.md`, convert a long instruction file into task-oriented "Start Here for ..." sections, add a concise "Must Follow" section with Always/Never rules, improve skill-discovery cues, or decide whether guidance belongs in root `AGENTS.md`, deeper docs, local `AGENTS.md`, or a repository-local skill.
Create or revise repository `ARCHITECTURE.md` files as short, stable architecture codemaps. Use when an agent needs to author or edit `ARCHITECTURE.md`, restructure an architecture document around a bird's-eye view and code map, clarify system boundaries, add architectural invariants or cross-cutting concerns, or decide which implementation detail should stay out of `ARCHITECTURE.md` and live in deeper design documents instead.
Create or revise Markdown `description` frontmatter so `docgarden match` routes agents to the right repository documents. Use when adding a new routed Markdown document, updating existing frontmatter, fixing noisy or missing match results, tuning skills or docs discovery, or making descriptions more query-aligned without loading document bodies.
Review the current branch, PR, or worktree against the plan or ExecPlan and record findings. Use when an agent needs a code review with findings for changes associated with an active plan.
Create, update, revise, reopen, or rescope a plan or ExecPlan using `PLANS.md`. Use when an agent needs to create a plan, plan a change, update the plan with new requirements, or rewrite an active ExecPlan.
| name | generator-execplan |
| description | Implement from an active plan or ExecPlan while keeping it up to date. Use when an agent needs to implement the active plan, continue plan-driven work, resume the current plan, or carry out an existing ExecPlan. |
| metadata | {"internal":true} |
Use this skill when the plan already exists and the task is to execute from it without losing the thread.
MUST READ before implementing:
docs/PLANS.mdPLANS.md defines the required ExecPlan structure, lifecycle, and maintenance rules.
As generator, you own:
You do not own:
completed/) or completion reviewUse the generator when the plan already exists and the task is to implement from it while keeping the living document current according to PLANS.md.
Update the ExecPlan as you go. Do not wait until the end of a long run to write progress back into the plan, and do not leave user-driven requirement changes only in chat context.
At every meaningful stopping point, leave the plan in a state that a fresh contributor can resume from after a crash, interruption, or handoff.
If requirements or success criteria change in a way that changes what the plan is asking for, stop and hand the plan back to planner-execplan before continuing.
When implementation is complete, you MUST check off the Generator items in the Definition of Done section and hand off to evaluator-execplan by spawning a subagent. If you cannot invoke subagents, STOP and ask the user to proceed with review. NEVER close the plan yourself.