agentleFS
OvichOvich/charrette/skills/execute/execute-plan/SKILL.md

execute-plan

Skill14 starsChanged 4 days agoMITView on GitHub

Use when carrying out an implementation plan the person has already approved, whether starting it or resuming it in a later session, and the tracker in it has to stay true while the work happens. Not for producing the plan (write-plan) and not for reviewing the result.

25 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use create-plan, to-spec and 6 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.
  • JetBrainsSkill

    create-plan

    Research the codebase and create an implementation plan with architecture notes, design document, and track decomposition. Use when starting a new feature or large change.

    Also in 4 projects that use this

  • GoogleSkill

    to-spec

    Turn the current conversation into a spec and publish it to the project issue tracker — no interview, just synthesis of what you've already discussed.

    Also in 3 projects that use this

  • MicrosoftSkill

    grill-with-docs

    Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. Use when user wants to stress-test a plan against their project's language and documented decisions.

    Also in 4 projects that use this

  • GoogleSkill

    grill-me

    A relentless interview to sharpen a plan or design.

    Also in 6 projects that use this

  • NVIDIASkill

    research

    Before designing a feature, research how comparable libraries solve the problem, check best practices, known pitfalls, and upstream changes. Use when starting a new feature, investigating an unfamiliar area, or the user asks to research something.

    Also in 4 projects that use this

  • Hugging FaceSkill

    implement

    Make tests pass. Invoke after /write-tests produces failing tests.

    Also in 3 projects that use this

  • +2 more from google

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 22 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Do not start: if the plan has no tracker, a phase has no goal, seats' roles and areas, or a check as a command (its tasks are written at …

    line 15
  2. Distinctive:

    Reconcile a tracker the repository has moved past: before starting; say what you found.

    line 16
  3. Distinctive:

    Stop for exactly three things:

    line 30

“Common” is measured: kept by at least 6 in 10 of the 22 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

Execute a plan

You are the plan's orchestrator; the orchestrate-plan skill holds those duties (the mandate, dispatch, checking the run, verification, merges, pull requests). Load it with this one.

The plan is a living document. write-plan built it and its tracker; this skill keeps both true as the work teaches something. Amending a task, adding one, dropping one or moving a decision is normal work, written into the plan as it happens.

Before starting

  • Open the plan in aiview every session and give the person the URL.
  • Do not start if the plan has no tracker, a phase has no goal, seats' roles and areas, or a check as a command (its tasks are written at its open). Say so and finish the plan first.
  • Reconcile a tracker the repository has moved past before starting; say what you found.
  • Every phase runs on the swarm board, however few seats it has, through the orchestrate-plan skill.

Keeping the plan and its tracker

The tracker's protocol is the write-plan skill's references/tracker.md, read before the first edit.

  • Tick as you go, never in a batch. ✅ only once this session has re-run the step's done-when.
  • Update before every handoff: a question, an approval, the end of a turn.
  • Change the plan before acting on the change. New work is a node, a dropped step is ✖ with the reason. A change that decides a module, interface, schema, contract or architecture is a row in the plan's decisions; one that becomes a slice gets its slice document first (write-slice).
  • A subagent's return is read, never pasted into a step; each line of it has a home, listed in the tracker's protocol.

When to pause

Stop for exactly three things:

  • A decision that is the person's. A finding that breaks a decision the plan rests on, a repair with real options, work outside the plan's scope.
  • Verification only a person can do. A browser flow, a real sign-in, another team's approval. Do everything mechanical first.
  • An action that leaves the machine and does not undo cheaply. A production deploy, a migration on shared data. Approved once each. A merge the mandate's pace gives the orchestrator is approved by the mandate.

A pause is a handoff: the tracker current; what happened, the decision wanted, the options and a recommendation; then what the person does next, numbered, each with its exact command. A secret goes in a file the pause names, by path and keys.

Do not pause for per-step sign-off, verification you can do yourself, naming, structure, tool choice, or a failure you understand whose fix changes nothing decided. Stop the line when one step fails twice for unrelated reasons.

The close

With a roadmap, say the plan finished and the roadmap may need redrawing; the person confirms the slot landed. When the plan delivered something a person looks at, recommend going over it with them through the validate-delivery skill.

List the decisions whose source is agent default that shipped, and ask whether any is to be revisited; when one of them is a contract, recommend the validate-delivery skill even if nothing was delivered for a person to look at.

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.