avn-plan-loop
Write a numbered plan with an acceptance check for each step, critique it once in a fresh pass, revise it, then execute one step at a time. Use when a task touches more than one file, has unclear scope, or failed before because steps were missed.
- License 许可
- MIT
- 所属仓库
- Aeven-AI/Aeven
- 最近更新
- 2026-11-15
- Stars
- ★ 1
- Forks
- 0
01 前言 Front Matter
- name
- avn-plan-loop
- description
- Write a numbered plan with an acceptance check for each step, critique it once in a fresh pass, revise it, then execute one step at a time. Use when a task touches more than one file, has unclear scope, or failed before because steps were missed.
02 资源文件 1
- 许可证LICENSE1.0 KB
03 SKILL.md 正文
avn-plan-loop
Plan before coding, check the plan once, then build the task one verified step at a time.
Steps
- Restate the goal in one sentence. List the files and commands you expect to touch. Read those files before you plan. Do not plan from memory.
- Write a numbered plan. Each step gets one action and one acceptance check that a command or a direct observation can confirm, for example "
npm test -- authpasses" or "GET /healthreturns 200". Keep each step small enough to verify in a single diff. - Write the plan to a file (PLAN.md, or the PROGRESS.md from avn-context-reset) so later steps and fresh sessions can read it.
- Critique the plan in a fresh pass. Reread only the goal and the plan, as a reviewer who did not write it. If you can, run the critique in a subagent or a new session. Answer three questions:
- Which step is missing (setup, migration, config, tests, cleanup)?
- Which assumption has not been checked against the code or docs?
- Is there a simpler path that reaches the same goal with fewer steps?
- Revise the plan. Check each risky assumption now by reading code or running a command. Do not carry it forward as a guess.
- Execute step 1 only. Run its acceptance check and read the real output. Use avn-verify-loop when the check fails.
- Mark the step done in the plan file, along with the evidence (command and result). Then move to the next step.
- If a step shows that the plan is wrong, stop and revise the remaining steps before continuing. Do not improvise around it.
Stop when
- Every step is marked done and its check passed, and the original goal holds when you check it end to end.
- You have done 2 critique rounds. A third round almost never finds anything new, so execute.
- A step fails its check twice in a row. Switch to avn-unstick.
Rules
- No step without a check. "Looks right" is not a check.
- Do one step per diff. Do not batch steps to save time.
- Do not add steps that the goal does not need.
- Keep the plan short: about 3-10 steps. Split bigger work into separate plans.
04 同仓库更多 Skills
cv
Make a CV or résumé with the user in one of Aeven's forty templates, from bold portrait-cover designs (Black, Slate, Fresh) to classic one-page resumes (Harvard, Tang, Minimal Rule), and deliver it as PDF, Word (DOCX) and Pages. Use for any CV, résumé or curriculum vitae request, when the user picked a template on the Resumes page ("…with the Black resume template"), or when they want an existing CV rebuilt in a new design.
cv
暂无简介
user-only-skill
Prove model-disabled skills stay outside the model catalog.
snapshot-skill
Exercise project skill discovery and loading in snapshot tests.
model-only-skill
Prove user-disabled skills remain available to the model.
editing-cordis-compositions
Use when creating, changing, or validating a Cordis composition for this harness — writing or editing an agent preset, adding or removing a plugin row, deciding whether something belongs to the host composition or to one session, checking whether a preset you authored actually mounts, or diagnosing a row that mounted but contributed nothing.