This file provides guidance to AI Agents when working with the powercat-canvas-apps plugin. A plugin for authoring Power Apps Canvas Apps. The Canvas Authoring MCP server (CanvasAuthoringMcpServer) exposes tools that agents use to generate, validate, and compile Canvas App YAML files (.pa.yaml) in conjunction with a running coauthoring studio session. The Power Apps Studio browser tab must remain open for the duration of the session — closing it ends the coauthoring session, which breaks compilecanvas and synccanvas operations. Skills orchestrate…
This file provides guidance to AI Agents when working with the powercat-code-apps plugin. A plugin for generating brand-aligned design guides for Power Apps code apps. It extracts brand tokens (colours, typography, spacing, layout patterns) from customer-provided assets and produces a design-guide.md that is consumed by the code-apps plugin (/create-code-app) to scaffold and build the app. Skills orchestrate specialist agents via the Task tool. Agents are not invoked directly by users. Test this plugin locally: The code-apps plugin from the official…
This file provides guidance to AI Agents when working with the powercat-dataverse plugin. A plugin for authoring and troubleshooting Microsoft Dataverse queries. It helps developers, makers, and architects write correct Web API URLs, convert FetchXML, target the right Power Apps surface, and diagnose common errors — without falling into the recurring traps that cause 90% of Dataverse Web API support threads. Skills orchestrate specialist agents via the Task tool. Agents are not invoked directly by users. Test this plugin locally:
This file provides guidance to AI Agents when working with the powercat-governance plugin. A plugin for Power Platform governance and administration. It helps Power Platform admins provision and manage environments consistently — replacing multi-step manual admin center workflows with a single prompt while enforcing organizational governance defaults. Skills orchestrate specialist agents via the Task tool. Agents are not invoked directly by users. Test this plugin locally: Spin up a fully-configured Power Platform Developer environment for any teammate in under 2…
This file provides guidance to AI Agents when working with the powercat-overflow plugin. A plugin for Power Platform makers, ALM owners, and CoE reviewers that audits every Power Automate cloud flow inside a Power Platform solution .zip against Microsoft's published coding guidelines. The skill unpacks the solution, walks each flow, classifies findings by impact (low / medium / high), writes a single [SolutionName].findings.json next to the uploaded ZIP, and opens the hosted PowerCAT-Overflow viewer (<https://microsoft.github.io/power-cat-skills/PowerCAT-Overflow.html>) with both the solution and…
This file provides guidance to AI Agents when working with the powercat-overpage plugin. A plugin for Power Pages makers, ALM owners, and CoE reviewers that audits a Power Pages site export (.zip produced by pac powerpages download, zipped) against Power Pages best practices across Security, Accessibility, Performance, Maintainability, SEO and Reliability. The agent unpacks the export, reads and reasons about the pages, templates, settings, web files and table permissions, writes a single [SolutionName].findings.json (validated against findings.schema.json) next to the uploaded…
Audience: Contributors modifying the core extension (extension) and shared packages (api, util). Emphasis: safe core evolution, API contract stewardship, and maximal reuse of @microsoft/vscode-fabric-util. Treat as public surface. Before changing an interface: Always check util/src before introducing new helpers: Commands (root orchestrates sequencing):
You are a triage-dispatch agent. Understand the user's request and delegate to the appropriate agent. You do not write code or make direct file changes. All code changes go through the orchestrator agent.
This section outlines the core implementation philosophy and guidelines for software development projects. It serves as a central reference for decision-making and development approach throughout the project. Embodies a Zen-like minimalism that values simplicity and clarity above all. This approach reflects: This development philosophy values clear documentation, readable code, and belief that good architecture emerges from simplicity rather than being imposed through complexity.
Compose context by including only the modules needed for the current task. Keep the context lean and relevant to maximize the model’s effectiveness during long sessions. This repo contains innovation…
Use AgentFactory for built-in agents or --agent-import-path for a custom BaseAgent subclass. ::: loopsbench.agents.agent_name ::: loopsbench.agents.base_agent ::: loopsbench.agents.agent_factory
* Review the CONTRIBUTING.md file for instructions to build and test the software. * Run the .github/Prime-ForCopilot.ps1 script (once) before running any dotnet or msbuild commands. If you see any build errors about not finding git objects or a shallow clone, it may be time to run this script again. IMPORTANT: This repository uses Microsoft.Testing.Platform (MTP v2) with xunit v3. Traditional --filter syntax does NOT work. Use the options below instead. Run all tests: Run tests for a specific test…
SBI (Secure Base Images) is a CLI tool that scans container base images and, in the default nightly workflow/config, performs vulnerability scanning of images from MCR (Microsoft Container Registry), then generates ranked recommendation reports. All commands use Task (see Taskfile.yml). Always run task lint before creating a PR — CI runs linting on all PRs and will fail on errors.
Purpose This repository follows Specification-Driven Development (SDD). Specifications are the source of truth. Code is an implementation of those specifications. AI agents are expected to behave as disciplined software engineers—not autonomous designers. When specifications are ambiguous, stop and ask for clarification rather than inventing behavior. This repository is intentionally specification-first. Development always follows the same lifecycle: Implementation is never the first step. The repository is divided into independent layers. Every layer depends only on layers above it. Never implement code…
You are working inside the samples/ folder of the Employee Self-Service Agent Developer Kit. This file is the authoritative briefing for any AI agent making changes here. For changes scoped to samples/, follow this file unless higher-priority system, security, or repository instructions say otherwise. Reference samples for the Microsoft Copilot Studio Employee Self-Service (ESS) agent. Each sample is a Copilot Studio topic plus its ESS Template Configuration XML(s). Samples are reference content — not a shipped product. See SUPPORT.md. Top-level…
If you are an AI coding agent (Copilot CLI, Claude, or any other) working in this directory tree, read this file first and follow it strictly. This document covers the rules for the production checks themselves (the code customers run via /flightcheck). Test-authoring rules are in tests/AGENTS.md. They complement each other; you need both. FlightCheck validates customer deployments by querying multiple APIs. Each API requires its own authentication and has different data available: Note: ../auth.py lives at scripts/auth.py, outside the…
If you are an AI coding agent (Copilot CLI, Claude, or any other) working on the test suite under this directory, read this file first and follow it strictly. This file is the test-authoring companion to solutions/ess-maker-skills/scripts/flightcheck/AGENTS.md. That sibling file covers the rules for the production checks themselves (the code customers run via /flightcheck). The cardinal rule applies in both places: Every FlightCheck check that calls an external API must be backed by a captured cassette of real responses AND…
Canonical location notice This repository follows Specification-Driven Development (SDD), and the AI Agent Operating Manual is a specification. To avoid two divergent constitutions, the single source of truth for agent behavior lives with the specifications, not in this build tree. Canonical file: dev-specs/ess-nextgen-migration-toolkit/AGENTS.md Read and follow the canonical AGENTS.md in full before implementing any task. Do not duplicate or fork its rules here. * Specifications (source of truth) live at the repository root under dev-specs/ess-nextgen-migration-toolkit/: * The buildable toolkit (implementation)…
You will help the user build out this React-based web app that visualizes data from Power BI semantic models. The app fetches live data via DAX queries, renders charts and grids using Vega-Lite and a built-in DataGrid component, and supports light/dark theming. Your job is to discover the user's data, write correct DAX queries, build React components that fetch and display that data, and validate the result in the browser. Break tasks into independent pieces and delegate them to sub-agents…
MicrosoftAGENTS.md