agentleFS
Sign inStart free
fabioc-alohafabioc-aloha/Alex_ACT_ONE/skills/project-capability-authoring/SKILL.md

project-capability-authoring

Skill0 starsChanged 9 days agoMITView on GitHub

Create tested project-local skills and scripts from demonstrated repeated work. Use after meditation or when a project workflow recurs and manual execution has caused inconsistency.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Do not use this for a one-off task, an untested aspiration, shared continuity, or a personal preference.

    line 20
  2. Very specific:

    Scan prior art: search existing .github/skills/, scripts/, and project guidance.

    line 27
  3. Distinctive:

    Validate: run the narrowest executable check for a script, and exercise the skill's should-fire and should-not-fire cases.

    line 43

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Project Capability Authoring

Turn a demonstrated repeated project workflow into a small local capability. This skill creates project-local skills and scripts. It does not change Core, build MCP servers, or write persistent memory.

When To Use

  • Meditation identifies a repeated workflow or deterministic task worth keeping.
  • A manual task has recurred and produced inconsistent results or avoidable mistakes.
  • The user asks to make a project workflow reusable, repeatable, or scripted.

Do not use this for a one-off task, an untested aspiration, shared continuity, or a personal preference.

Author A Capability

  1. Prove the pattern: name the repeated workflow, the concrete failure or inconsistency it prevents, and the project users who benefit.
  2. Scan prior art: search existing .github/skills/, scripts/, and project guidance. Extend an existing capability when it already covers the workflow.

  3. Choose the smallest placement:

    NeedDestination
    Guided repeatable workflow.github/skills/<name>/SKILL.md
    Workflow with deterministic automation.github/skills/<name>/scripts/<task>.cjs
    Cross-cutting project automationscripts/<task>.cjs
  4. Use the shared authoring owner: load and follow compile-brain for the explicitly selected draft, its contract, semantic preservation, and validation. This does not bypass the proof-of-need, local-scope, or approval requirements here. Define the contract: write the trigger, inputs, outputs, failure mode, and one should-fire plus one should-not-fire example. A script must state its prerequisites, accepted arguments, and validation command.
  5. Preview the change: show every proposed file and the validation plan. Obtain explicit user approval before creating or modifying project files.
  6. Validate: run the narrowest executable check for a script, and exercise the skill's should-fire and should-not-fire cases. Run diagnostics on every changed file.
  7. Report: state the extracted pattern, artifact path, validation evidence, and why a new capability was warranted instead of a one-off solution.

Memory Boundary

Meditation handles the project's AGENTS, HANDOFF, task list, and changelog; it does not write episodic or native memory. This skill owns only separately approved local capabilities. Do not create a chronicle or memory record as an authoring receipt; report evidence through the project's existing change and validation records.

Broadly reusable or Mall-worthy candidates remain recommendations for separate review. Local authoring approval does not authorize global installation, marketplace submission, or edits to another repository. Apply the PII filter before including any project evidence in a proposed reusable capability.

Anti-Patterns

Anti-patternCorrection
Creating a skill after one routine taskRequire a demonstrated repeat or concrete failure first.
Copying a skill that already exists locallyExtend the existing capability after the prior-art scan.
Hiding a destructive script behind a skillPreview the command and require explicit approval before execution.
Creating project files while summarizing meditationMeditation routes the candidate; this skill previews and validates it.
Treating an MCP server as a project scriptEvaluate MCP construction separately; it is not this skill's scope.

Would Revise If

Revisit if real meditations produce no demonstrated reusable workflow, or authors repeatedly choose the wrong local placement.

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.

Check your setup, free