How real projects brief Codex, Cursor and every other agent that reads AGENTS.md.
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.
MicrosoftAGENTS.mdmicrosoft/experiment-catalog/.agents/skills/harness-engineering-playbook/assets/templates/AGENTS.md
Replace placeholder text and keep this file compact, command-first, and operational. Run from repository root:
MicrosoftAGENTS.mdmicrosoft/github-advisory-mcp/AGENTS.md
Guidance for AI coding agents (and humans) working in this repository. Keep changes minimal, typed, and covered by tests. For contribution policy see CONTRIBUTING.md; for user-facing docs see README.md. An MCP server that serves GitHub Security Advisories from a local clone of github/advisory-database. Two-tier design: MCP server (stdio or HTTP) → local Express REST API → LocalRepositoryDataSource (reads advisory JSON from disk). Core MCP tools: list_advisories, get_advisory (full parameters in the README). The server reads advisory JSON from ADVISORY_REPO_PATH (default…
MicrosoftAGENTS.mdmicrosoft/microsoft-foundry-e2e-js/AGENTS.md
This repository includes custom GitHub Copilot agents to help learners navigate the workshop and troubleshoot issues. This workshop teaches developers how to build AI agents end-to-end on Microsoft Foundry. Each task (02–09) has three components: In GitHub Copilot Chat (VS Code or Codespaces), mention the agent by name: Each agent is scoped to a specific learning path and will only provide guidance for that path. Pick the agent that matches how you want to work through the quest. When providing…
MicrosoftAGENTS.mdmicrosoft/PurviewIntegrations/AGENTS.md
This file is the single entry point for AI agents (Copilot CLI, Copilot Chat, Claude, Cursor, etc.) working in this repository. It points to the tooling, conventions, and build/test loop an agent needs to make a change and validate it before pushing. Purview GitHub Action — a TypeScript GitHub Action that scans repository files on push / pull request and sends them to Microsoft Purview for data governance and compliance tracking. It authenticates with Azure via GitHub OIDC, processes changed…
MicrosoftAGENTS.mdmicrosoft/Tandem/development/AGENTS.md
These instructions apply within development/. All paths below are relative to this directory, which retains the original source layout. The approved Node integration composes this engine with root src/ through human-loop-controller.mjs. Root build/pack scripts may package this complete subtree, and root checking may emit its type declarations; keep its own package, lockfile and strict build/tests. Root package.json alone owns the integrated release version. Do not activate the nested discovery shim or overwrite a user installation without authorization. Use actual Node…
MicrosoftAGENTS.mdmicrosoft/frontier-agentic-cobuild-rvas/AGENTS.md
Before drafting, editing, or returning repository prose, invoke humanize-content. This includes documentation, Markdown, and user-facing explanations. Do not apply it to files under plugins/ or to AGENTS.md. Use TONE-OF-VOICE.md unless the user requests another voice. Write for competent engineers: lead with the point, use concrete verbs, name the actor when it matters, and explain complexity only when the reader needs it. Preserve technical facts, Microsoft product names, dates, citations, and governance terms. Apply diminishing returns: keep material only when it…
MicrosoftAGENTS.mdmicrosoft/trivia-challenge/AGENTS.md
Preserve gameplay behavior unless a change is requested. Before changing gameplay, authentication, SQL persistence, or Function contracts, read Architecture. Consult src/config/gameConfig.ts for executable settings. For Rayfin changes, load .agents/skills/rayfin/SKILL.md and read the installed package documentation identified by rayfinDocs. CLI documentation is in @microsoft/rayfin-guide. Preserve explicit private entity policies and schema discovery; run npm run schema:check after changing them. For deployment or environment changes, read Deployment. Deployment and destructive cloud operations require explicit authorization. Local gameplay is unsupported, and npm run…
MicrosoftAGENTS.mdmicrosoft/agentic-loops/AGENTS.md
Go to .github/copilot-instructions.md.
MicrosoftAGENTS.mdmicrosoft/Foundry-Agent-Solution-Templates/apim-hosted-agent/AGENTS.md
This project is a Microsoft Foundry hosted agent fronted by Azure API Management (APIM). Preserve the governed request paths: callers enter through the APIM agent API, the hosted agent reaches its model through the APIM model API, and MCP tools are exposed through APIM tool APIs. This project was built with the microsoft-foundry skill. Before working on or answering questions about Foundry agents, read the microsoft-foundry skill first. - azure.yaml — azd orchestration for the Foundry project, toolbox, hosted agent,…
MicrosoftAGENTS.mdmicrosoft/Foundry-Agent-Solution-Templates/enterprise-knowledge-agent/AGENTS.md
Preserve the architecture: one Responses-protocol Hosted Agent connects only to one Foundry Toolbox. The Toolbox exposes the composed Foundry IQ Knowledge Base and web_search; the agent must not initialize Search or web clients directly. Keep Search keyless with local authentication disabled. Keep 2026-08-01-preview pinned for all Knowledge Source and Knowledge Base operations. Configuration fragments are native platform payloads and must not contain plaintext secrets. Terraform and azd are the default deployment path; Bicep remains a supported companion implementation. Keep both…
MicrosoftAGENTS.mdmicrosoft/Foundry-Agent-Solution-Templates/private-network-hosted-agent/AGENTS.md
Preserve these invariants: - Terraform and azd are the default deployment path. Bicep remains a supported companion implementation. Keep both implementations behaviorally equivalent. - The default Hosted Agent is Python 3.13 source code with remote_build. - The opt-in ACR scenario consumes only a digest-pinned image from an existing enterprise ACR and creates only its Foundry project connection. - Public fallback, Dockerfiles, template-owned ACR resources, ACR networking, ACR IAM, App Insights, and Log Analytics are out of scope. - Do not…
MicrosoftAGENTS.mdmicrosoft/agentic-journeys/AGENTS.md
This repository contains agentic journeys that build and deploy applications to Azure using GitHub Copilot agents, skills, and Azure Developer CLI (azd). Journeys are self-contained — learners can start with any of them. The root README.md → Agentic journeys table lists them all with cost estimates. Plugin (canonical — use everywhere): Do not use alternate marketplace names. Install the Azure Skills plugin for access to Azure-specific MCP and skills tools (Bicep schemas, deployment planning, architecture diagrams, log analysis): Once inside…
MicrosoftAGENTS.mdmicrosoft/amplifier-bundle-context-intelligence/AGENTS.md
Guidance for AI agents and developers working in this bundle repository. Use Foundation validate-bundle-repo v3.16.1 or later. Earlier validators can mistake glob/template storage paths ending in */context-intelligence or {id}/context-intelligence for an invocation of the internal mode. The fix belongs in Foundation's mode-reference matcher, not this bundle's storage paths or mode visibility. Keep modes/context-intelligence.md at advertised: false. Do not delete valid path references, advertise the internal mode, or treat the resulting ERROR as a PASS. Select the corrected recipe with CI_VALIDATE_RECIPE…
MicrosoftAGENTS.mdmicrosoft/amplifier-context-intelligence/AGENTS.md
Event-driven telemetry platform for Amplifier sessions. Captures session events as structured data and builds a property graph in Neo4j. See README.md for full setup instructions. - The shipping server Dockerfile uses an approved MCR Azure Linux base to meet S360 / Qualys SCA requirements. Do not switch it to an unapproved base or add production containers on unapproved bases; test containers are fine. - Preserve .dockerignore and docker-entrypoint.sh with the server image build and first-run flow. Before editing the Dockerfile,…
MicrosoftAGENTS.mdmicrosoft/amplifierd/AGENTS.md
This codebase is yours: what it's for and how it's built are your call, and you'll keep building on it.
MicrosoftAGENTS.mdmicrosoft/amplifier-docs/docs/user_guide/agents.md
Specialized sub-agents for focused tasks
MicrosoftAGENTS.mdmicrosoft/amplifier-unified/AGENTS.md
This codebase is yours: what it's for and how it's built are your call, and you'll keep building on it. Before broad source reads or searches, use the source and ownership map. python scripts/source_context.py summarizes tracked inputs without reading their bodies; use --format paths with --under to select relevant owners, contracts and tests. Ordinary rg does not apply these context classifications; use the map's selective-search examples, and expand to retained evidence or generated output when the task needs it. Use…
MicrosoftAGENTS.mdmicrosoft/amplifier-unified/components/AGENTS.md
components/workspaces/ is the temporary authoring home for the independently buildable amplifier_workspace_metadata package. Its registration and query mechanism tests belong here; Unified retains initialization, filesystem effects, history, authority and persistence. Unified consumes only the published artifact selected into workspace-metadata-delivery/ by workspace-metadata.source.json, as the sole installed owner; do not coinstall the standalone distribution. Never edit that generated copy by hand. Author edits and publication do not refresh the app selection. This metadata boundary is not the whole workspace domain. components/outputs/ is temporarily…
MicrosoftAGENTS.mdmicrosoft/amplifier-unified/components/web-client/AGENTS.md
This codebase is yours: what it's for and how it's built are your call, and you'll keep building on it. Read README for the current HTTP/SSE boundary, pinned upstream vision/contracts and build/test scopes. The repository is private; preserve its visibility. Unified selects a client revision and qualifies its packaged pairing; source work here does not publish or adopt a new default automatically. Author browser behavior in src/, standalone pages in public/, client checks in tests/, and build/preparation tools in scripts/.…
MicrosoftAGENTS.mdmicrosoft/capacity-toolkit/AGENTS.md
Instructions for an AI agent (e.g. GitHub Copilot CLI) driving this read-only-by-default toolkit against an Azure tenant. If you are an agent: read this file fully before running anything. This toolkit is READ-ONLY BY DEFAULT. You exist to observe and report. It contains exactly one write tool — Deploy-QuotaGroups.ps1, which provisions Azure Quota Groups — and you must treat it as off by default: never run it unless the human explicitly asks you to provision quota groups in this run…
An open format for instructions to coding agents, read by Codex, Cursor and others. Think of it as a README written for agents.
At the repository root, with more specific files in subdirectories. Agents read the one closest to the file they're editing.
Setup and test commands, code style, and the rules a new contributor would need to know.
Claude Code reads CLAUDE.md. A one-line CLAUDE.md that points at AGENTS.md covers both.
