agentleFS
Sign inStart free
ryanportfolioryanportfolio/Harness-Firmware/.agents/skills/writing-plans/SKILL.md

writing-plans

Skill1 starsChanged 21 days agoMITView on GitHub

Use when a clear task needs a multi-step implementation plan, dependency ordering, or a durable handoff. Skip for routine changes that can be executed directly.

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 172 edited versions we read keep these.

  • 5

    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:

    Read the relevant code and repository instructions before naming files or commands.

    line 12
  2. Distinctive:

    When planning extraction or migration of repeated operations, read selective shared-code refactoring.

    line 17
  3. Distinctive:

    Before execution, check scope coverage, dependencies, failure paths, and final acceptance.

    line 31
  4. Distinctive:

    Update the plan when evidence or user requirements change.

    line 32
  5. Distinctive:

    Do not ask the user to choose an execution mechanism unless it changes a meaningful cost, constraint, or outcome.

    line 35

“Common” is measured: kept by at least 6 in 10 of the 172 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 implementation

Turn the accepted goal into an executable sequence. Resolve material missing requirements first; do not reopen choices the user already made. A request for a plan ends with a plan. A request to implement continues into authorized work once the plan is ready.

Read the relevant code and repository instructions before naming files or commands. Use an exposed planning tool if available; otherwise keep a short Markdown checklist. Save a plan in the repository's preferred location when the user requests a document, a handoff needs one, or the work must survive a context reset. Otherwise keep planning lightweight.

When planning extraction or migration of repeated operations, read selective shared-code refactoring.

Each step identifies:

  • The observable outcome and affected files or interfaces.
  • Prerequisites and unresolved decisions that could change the approach.
  • A check that establishes completion at the claimed layer.

Order by dependencies and deliver a thin working path (a walking skeleton) early. Make steps small enough to verify independently. Distinguish local step checks from final integration acceptance. Include code or exact commands where they remove ambiguity, not a second implementation of the entire solution. Discover verifiable facts instead of filling the plan with placeholders.

Before execution, check scope coverage, dependencies, failure paths, and final acceptance. Update the plan when evidence or user requirements change; preserve the reason for material changes. Record blocked checks without marking them complete. Continue independent work.

Do not ask the user to choose an execution mechanism unless it changes a meaningful cost, constraint, or outcome. Planning alone does not authorize delegation, package installation, commit, push, merge, deployment, or a new sidebar task. Follow current authorization for those actions. At handoff, include the workspace, current state, next action, and remaining checks so another agent can continue without this conversation.

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.

Reviews can't be read right now.

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.

Check your setup, free