Microsoft Agent 365 Skills is a collection of coding-agent skills and MCP configuration for building and operating AI Teammates hosted in Microsoft Teams. It supports the Agent 365 lifecycle, including WorkIQ integration, registration, observability, validation, and local testing, for users of Claude Code and GitHub Copilot.
Borrowing it
Nothing to install: this file belongs to microsoft/agent365-skills. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/microsoft/agent365-skills/main/.github/copilot-instructions.mdgit clone --depth 1 https://github.com/microsoft/agent365-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/microsoft/agent365-skills/copilot-instructions)<a href="https://agentmods.dev/instructions/microsoft/agent365-skills/copilot-instructions"><img src="https://agentmods.dev/badge/instructions/microsoft/agent365-skills/copilot-instructions/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/microsoft/agent365-skills/copilot-instructions"><img src="https://agentmods.dev/badge/instructions/microsoft/agent365-skills/copilot-instructions.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.11406 | $0.11406 |
| Opus 5.5 | $0.04562 | $0.04562 |
| Sonnet 5.5 | $0.02281 | $0.02281 |
| Haiku 4.5 | $0.01141 | $0.01141 |
Grade A, and why
agent365-skills copilot-instructions.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 20d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 426 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent 365 Skills — GitHub Copilot Instructions
Skills for instrumenting and registering Microsoft Agent 365 agents. When a user asks for any of the trigger phrases below, follow the corresponding SKILL.md exactly.
Quick reference
| Skill | When to invoke | Delegates to |
|---|---|---|
a365-setup |
Entry point — CLI install, Azure prereqs, capability menu | make-ai-teammate (AI Teammate path) or make-a365-agent (other paths) |
make-ai-teammate |
Transform an agent into an AI Teammate (Teams / Copilot publish) | instrument-observability (auto), add-workiq-tools (offered) |
make-a365-agent |
Non-AI-Teammate blueprint provisioning (Register / Observability / WorkIQ paths) | instrument-observability (optional), add-workiq-tools (optional, skipped for S2S) |
add-workiq-tools |
Wire MCP servers (Mail / Calendar / Word / etc.) into the agent | — |
instrument-observability |
OTel + A365 tracing exporter wiring | — |
a365-code-validator |
A365 observability/MAC Activity validation with optional guided fixes | — |
purview-dlp-integration |
Add a Purview DLP gate that blocks sensitive prompts before the LLM | — |
test-local |
Launch agent + AgentsPlayground for local smoke test | — |
Skills are designed to be additive, idempotent, and state-aware — re-running is safe.
Task Execution Discipline
When a skill creates a task list, run every task to completion in one turn — do not pause
between phases. Mark each task complete the moment its phase finishes (TaskUpdate in
Claude Code, - [ ] → - [x] in Copilot Chat / CLI). Surface a one-line status update
between phases instead of asking permission.
Only pause at these explicit interaction points:
a365-setup: capabilities menu; authMode (non-AI-Teammate only); install confirmations for missing tools; Azure login.make-ai-teammate: Phase 0B confirm; 0C row-8 sub-question; 9.6 WorkIQ offer; 9.7.1a Reuse/Re-run/Fresh; 9.7.2 Run Target + 9.7.2b hosting.make-a365-agent: agent name + directory; reuse-blueprint; hosted vs local; observability/WorkIQ offers.add-workiq-tools: MCP server selection; Word @mention offer (only whenmcp_WordServeris selected and stack is Node.js LangChain).instrument-observability: agent kind + auth mode (only if not in cache).a365-code-validator: after the report, asks whether to apply safe fixes, create a fix plan, or stop.purview-dlp-integration: DLP policy choice (new / existing / skip); agentic auth handler name and app id/display name only when not discoverable from a365 config.test-local: confirm before launching.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 20d ago Changed · +3 tokens per session c03fa675e415
- 1mo ago Changed · +591 tokens per session 804308999911
- 1mo ago Changed · +46 lines · +1,370 tokens per session 4b257ce52cb8
- 2mo ago First seen · 380 lines · 9,442 tokens per session scan A ffc6a92eeaf0
agent365-skills copilot-instructions.md is an instructions file published in the GitHub repository microsoft/agent365-skills (50 stars, last pushed 16d ago), licensed MIT. It adds 11,406 tokens to every session, about $0.0456 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
agent365-skills copilot-instructions.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| agent365-skills copilot-instructions.mdmicrosoft/agent365-skills | Instructions file | 50 | 16d ago | MIT | A | 11,406 | GitHub Copilot |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | today | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |