agentleFS
matthewwong525matthewwong525/WongStack/.agents/skills/plan/SKILL.md

plan

Skill7 starsChanged 11 days agoMITView on GitHub

Draft a change's plan and review page, or a to-do for work changing no repo file; also takes pasted review notes.

184 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use heartmula, webhook-subscriptions and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • math-incSkill

    heartmula

    Set up and run HeartMuLa, the open-source music generation model family (Suno-like). Generates full songs from lyrics + tags with multilingual support.

    Also in 9 projects that use this

  • Tommy-ywSkill

    webhook-subscriptions

    Create and manage webhook subscriptions for event-driven agent activation, or for direct push notifications (zero LLM cost). Use when the user wants external services to trigger agent runs OR push notifications to chats.

    Also in 5 projects that use this

  • math-incSkill

    minecraft-modpack-server

    Set up a modded Minecraft server from a CurseForge/Modrinth server pack zip. Covers NeoForge/Forge install, Java version, JVM tuning, firewall, LAN config, backups, and launch scripts.

    Also in 5 projects that use this

  • john-data-chenSkill

    touchdesigner-mcp

    Control a running TouchDesigner instance via twozero MCP — create operators, set parameters, wire connections, execute Python, build real-time visuals. 36 native tools.

    Also in 6 projects that use this

  • gaebalaiSkill

    codex-system

    PROACTIVELY consult Codex CLI, your highly capable supporter with exceptional reasoning and task completion abilities. Codex is a trusted expert you should ALWAYS consult BEFORE making decisions on: design choices, implementation approaches, debugging strategies, refactoring plans, or any non-trivial problem. When uncertain, consult Codex. Don't hesitate - Codex provides better analysis. Explicit triggers: "think deeper", "analyze", "second opinion", "consult codex".

    Also in 3 projects that use this

  • ibahgatSkill

    context-optimize

    Improve context efficiency and cache stability for long or complex sessions.

    Also in 3 projects that use this

  • +3 more from math-inc

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 1 is distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 167 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Create an apply-ready OpenSpec change and its review.html page.

    line 9
  2. Very specific:

    Run /explore in bounded mode.

    line 17
  3. Very specific:

    Write every artifact in the planning set by the CLI contract, in the selected root or store.

    line 23
  4. Distinctive:

    Before tasks, weigh deterministic code for a repeated process (the principles).

    line 25
  5. Very specific:

    Run step-title.mjs set PLAN --task "<proposal title>".

    line 55

“Common” is measured: kept by at least 6 in 10 of the 167 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

/plan

Create an apply-ready OpenSpec change and its review.html page. First run node .claude/skills/explore/scripts/step-title.mjs set PLAN --task "<the work>".

Change only plans and wiki pages, and save a schedule to the app after an OK; code waits for the person's go (limits).

Work that changes no repo file gets no change, page, or file: after the bounded /explore pass, write a short numbered to-do in chat, marking each step that acts outside it (outward) (verbs for any work).

Explore first

Run /explore in bounded mode; it asks by the exit round. Log each decision in the Decision log: **YYYY-MM-DD** — Asked <question> → chose <answer>. for every answer, **YYYY-MM-DD** — Assumed: <decision>, because <reason>. for each inferred one.

New workspaces chosen → open one per other part, report before drafting, and plan only this chat’s part.

Draft with the CLI

Write every artifact in the planning set by the CLI contract, in the selected root or store. Complete in place an incomplete change /apply selected; a block goes back to /apply, unbuilt.

Before tasks, weigh deterministic code for a repeated process (the principles); raise it in the exit round if it changes scope. Testable behavior, not prose, gets test authoring beside source tasks.

Why and What Changes use plain words: what the person will see or do. Deleting/reshaping data, sending a message, or removing a key gets one line: it can't be undone; otherwise none. Put file names, code and commands in design/specs/tasks, hidden on the page. Capabilities and Impact may be technical.

Draw in the proposal, then build the page

Draw as you write, with no second agent: one drawing per What Changes bullet by default, as a fence indented in its bullet by the drawing guide. A card shows the closest real result: a screen change gets a prototype, a site this app doesn't run a live one; notes update it too. Every plan lists what the build needs.

Build the page after drafting and each edit:

node "$(git rev-parse --show-toplevel)/.claude/skills/plan/scripts/build-review.mjs" "<change-root>" --require-current --link

Copy its status, plan copy, link, and review lines as printed; print the plan's link says when the last goes. Fix warned drawings. Keep no other copy of the proposal's text: a second copy goes stale.

A change that adds or restructures a screen gets a ## UX design section; sketch phone work phone-first.

Review notes

Notes on the plan <name> from the review page (also Update the plan <name> with these notes) updates its existing plan without an explore round. A chat message about a waiting plan is handled the same way, after a bounded explore pass on only what it opens. Get paths from openspec status --change "<name>" --json (contract); a missing change → report and stop. Trust quoted text over spot labels.

  • A question note gets an answer, no edit or log entry; offer suggested edits only when standalone.
  • A change note is applied coherently, each change or decline logged; validate, create only requested artifacts, rebuild and print the page's link.
  • Notes alone finish without building. Sent with an explicit build or publish choice, they return to /apply or /ship, which goes on unasked: the person already chose. A material blocker stops.

Finish

Lay out tasks so the build is written before it is checked (why): group source and test authoring by surface, and put required checks in one final phase, not per-part /save tasks. A substantive deployed acceptance task names /save after implementation.

Run step-title.mjs set PLAN --task "<proposal title>". Validate with openspec validate "<name>" --strict --no-interactive; confirm the review page and every apply-required artifact are complete. Standalone, end on the plan's copy, link, and review line, asking nothing. Invoked by /apply, return the exact change name to it.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.