菜单
Write a detailed, executable plan for a single unit of work, then get it approved by review-plan. Updates work/<slug>/README.md with concrete instructions and sets status to ready (or needs-input if review repeatedly rejects).
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Show the user the agent's work on a research project and save iterations on the user's behalf. Scaffold rendering and deploy infrastructure (Quarto today, GitHub Pages, dev container), show the rendered output, save iterations. Doesn't handle research execution (use `asta-flows`).
Install or update the `asta` CLI to the version pinned by this plugin. Invoke whenever `asta` is not on PATH (e.g. `command not found` for `asta`) or when a subcommand/flag appears missing — that signals a stale install needing an update.
Converse with the user to identify the next most productive units of work for the current project, and record them in project.md under Pending Work. Use when the user wants to plan what to work on next, or when there is no project.md yet.
Execute the approved plan in work/<slug>/README.md (status ready), write outputs into work/<slug>/data/, record what happened in the Results section, then hand off to review-work. Use when there is a ready work item.
Critic skill. Given a plan in work/<slug>/README.md (status pending-review), check that the steps are concrete, the tools are appropriate, and the plan is realistic. Write a verdict and feedback back into the README.
Critic skill. Given an executed work item (status pending-assessment), assess whether the Goal was actually accomplished. Identify the root cause when it was not (incorrect assumption, technical blocker, missing data). Updates work/<slug>/README.md Assessment section.
| name | plan-work |
| description | Write a detailed, executable plan for a single unit of work, then get it approved by review-plan. Updates work/<slug>/README.md with concrete instructions and sets status to ready (or needs-input if review repeatedly rejects). |
| allowed-tools | Read(work/**) Read(project.md) Edit(work/**) Write(work/**) Skill(asta-assistant:review-plan) |
Takes a vague goal in work/<slug>/README.md and produces a concrete, tool-aware plan that a downstream agent can execute without further design decisions.
work/<slug>/README.md with at least the Goal section filled in. Frontmatter status is usually pending-plan.project.md for surrounding context (overall goal, background, related completed work).Same work/<slug>/README.md with:
# Instructions populated with a concrete plan (numbered steps; specific tools, commands, files, or skills to invoke; expected artifacts).status: ready (on approval) or status: needs-input (after repeated rejections).plan_review_attempts: <int> to track loops.Read context. Open the work README and project.md. If the README's Goal section is empty or unclear, set status: needs-input and stop — do not invent a goal.
Draft the plan. Under # Instructions, write numbered steps that:
work/<slug>/data/<file>).Update status and counter.
status: pending-review
plan_review_attempts: <n+1>
Invoke review-plan with the slug. The review writes its verdict into the README (see review-plan).
Read the verdict.
status: ready — done.status: pending-plan and the review left feedback in the README, incorporate it and loop to step 2.status: needs-input, leave the latest review feedback in place, and stop. Surface to the user.A good plan is one a fresh agent could execute without asking design questions. Concretely: