agentleFS
MicrosoftKnown publishermicrosoft/power-cat-skills/plugins/powercat-pcf/AGENTS.md

power-cat-skills / powercat-pcf

AGENTS.md70 starsChanged 2 days agoMITView on GitHub

This plugin authors virtual Power Apps Component Framework controls for model-driven apps. There is no plugin build step: the deliverables are skill instructions and reference documentation. Validate JSON metadata, skill frontmatter, resource links, and marketplace registration when editing this plugin. Write references independently from official documentation, cite the relevant sources, and distinguish platform guarantees from project policies. Do not borrow third-party skill text, examples, or organization. Generated PCF projects must use their own build, lint, and tests plus verification in…

The file

We split this AGENTS.md into the rules it gives an agent

6 rules. About 0 are common, 2 are distinctive, and 4 are 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.

  • 4

    Very specific

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

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

    Always scaffold new controls using pac pcf init --framework react: field for a single column, dataset for views/subgrids.

    line 16
  2. Very specific:

    Use ComponentFramework.ReactControl and control-type="virtual".

    line 18
  3. Distinctive:

    Verify current documented platform-library versions before choosing dependencies.

    line 20
  4. Very specific:

    Use bound outputs for form field saves, the bound dataset for view/subgrid reads, and context.webAPI for Dataverse operations with the …

    line 25
  5. Very specific:

    Always exercise the local harness with npm start watch.

    line 26
  6. Distinctive:

    Write references independently from official documentation, cite the relevant sources, and distinguish platform guarantees from project …

    line 34

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md 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
AGENTS.md

Power CAT PCF Plugin

This plugin authors virtual Power Apps Component Framework controls for model-driven apps.

Structure

  • .claude-plugin/plugin.json - Plugin metadata.
  • skills/create-pcf/SKILL.md - Requirements, implementation, validation, and delivery workflow.
  • skills/create-pcf/references/quality-checklist.md - Cross-cutting acceptance tests.
  • skills/create-pcf/references/component-patterns.md - Field/dataset interaction design.
  • skills/create-pcf/references/manifest-reference.md - Model-driven manifest authoring contracts.
  • skills/create-pcf/references/control-lifecycle.md - Virtual lifecycle, state reconciliation, and cleanup.

Invariants

  • Always scaffold new controls using pac pcf init --framework react: field for a single column, dataset for views/subgrids. Preserve an existing virtual scaffold when extending it.
  • Author TypeScript only (.ts/.tsx), never plain JavaScript. Never edit auto-generated types, manifests, bundles, or solution/build outputs; edit their authored inputs and regenerate with tooling.
  • Use ComponentFramework.ReactControl and control-type="virtual".
  • Declare React and Fluent v9 platform libraries. Do not bundle their implementations or introduce Fluent v8.
  • Verify current documented platform-library versions before choosing dependencies. Runtime versions are not manifest compatibility targets.
  • For non-text field controls, show --- for null and select an agreed default only on deliberate activation. Focus, rendering, and loading must never write a default.
  • Preserve valid falsy values, clear-to-null behavior, host updates, disabled/read-only states, and field security.
  • Readable read-only fields/cells show only a human-readable value label or static ---, never disabled editor UX. Hide editing affordances, cancel uncommitted drafts on mode changes without writes, and never expose unreadable/secured data.
  • Accessibility and keyboard behavior are acceptance criteria, not optional polish.
  • Use bound outputs for form field saves, the bound dataset for view/subgrid reads, and context.webAPI for Dataverse operations with the required manifest feature declaration.
  • Always exercise the local harness with npm start watch; unsupported WebAPI/dataset/security behavior requires separate host verification.
  • Package releases with pac solution init, pac solution add-reference, and msbuild /t:build /restore, explicitly selecting production and managed/unmanaged output. Deploy approved, verified release artifacts with pac solution import.
  • pac pcf push is optional only for approved inner-loop development deployments; never use it for downstream test/UAT/production or as proof of release readiness.

Development

There is no plugin build step: the deliverables are skill instructions and reference documentation. Validate JSON metadata, skill frontmatter, resource links, and marketplace registration when editing this plugin.

Write references independently from official documentation, cite the relevant sources, and distinguish platform guarantees from project policies. Do not borrow third-party skill text, examples, or organization.

Generated PCF projects must use their own build, lint, and tests plus verification in a model-driven app. Do not equate the local harness with host verification.

Use the consumer's configured npm registry for all npm operations and honor their organization policies. Never override the registry or bypass a configured proxy. If a requested version is unavailable, query that same registry and pin the newest compatible available version, or report the blocker. This plugin does not require a Microsoft-internal feed.

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.