Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/microsoft/agent-governance-toolkit/copilot-instructionsgit clone --depth 1 https://github.com/microsoft/agent-governance-toolkitWhat 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 | $0.05833 | $0.05833 |
| Opus 5 | $0.02916 | $0.02916 |
| Sonnet 5 | $0.01167 | $0.01167 |
| Haiku 4.5 | $0.00583 | $0.00583 |
Grade A, and why
agent-governance-toolkit 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 yesterday.
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 — 408 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Copilot Instructions for agent-governance-toolkit
Architecture
Multi-language monorepo for runtime governance of AI agents. Core is Python, with SDKs in TypeScript, Rust, .NET, and Go.
Core Python packages live under packages/:
| Package | Purpose |
|---|---|
agent-compliance (agent_governance_toolkit) |
Unified installer, agt CLI, runtime policy enforcement |
agent-os (agent_os_kernel) |
Kernel: policy evaluator, identity, trust, audit |
agent-mesh (agentmesh_platform) |
Multi-agent networking: Wire Protocol, registry, relay, E2E encryption |
agent-hypervisor |
Execution sandboxing and resource isolation |
agent-sre |
SRE: health checks, circuit breakers, observability |
agent-runtime |
Runtime orchestration layer |
agent-mcp-governance |
MCP (Model Context Protocol) governance primitives |
agent-discovery |
Agent discovery and capability advertisement |
agent-lightning |
Lightweight/fast-path policy evaluation |
agent-marketplace |
Agent marketplace and registry |
agentmesh-integrations |
Framework adapters (LangChain, CrewAI, AutoGen, etc.) |
Language SDKs:
| Language | Path | Package name |
|---|---|---|
| TypeScript | packages/agent-mesh/sdks/typescript/ |
@microsoft/agent-governance-sdk |
| Rust | packages/agent-mesh/sdks/rust/ |
agent-governance, agent-governance-mcp |
| .NET | agent-governance-dotnet/ |
Microsoft.AgentGovernance |
| Go | agent-governance-golang/ |
(standalone top-level) |
Other top-level directories: docs/ (MkDocs site), examples/ (runnable integrations), demo/ (live dashboards), pipelines/ (ESRP release automation), benchmarks/, fuzz/ (ClusterFuzzLite), notebooks/.
Each subdirectory with an AGENTS.md has area-specific commands and boundaries: read it before changing code there.
Build, Test, and Lint
Python
# Install (editable, all extras)
pip install -e ".[full]"
# Lint
ruff check --select E,F,W --ignore E501
# Test (full suite)
pytest
# Test (single file)
pytest tests/test_policy.py
# Test (single test by name)
pytest -k "test_policy_deny_dangerous_tools"
# Build
python -m build
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.
- yesterday First seen · 408 lines · 5,833 tokens per session scan A d051c986f5a4
agent-governance-toolkit copilot-instructions.md is an instructions file published in the GitHub repository microsoft/agent-governance-toolkit (6,153 stars, last pushed 6d ago), licensed MIT. It adds 5,833 tokens to every session, about $0.0292 per session on Opus 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.
Other instructions, from other repositories
orloj AGENTS.md
Instructions for OrlojHQ/orloj, covering orloj agent instructions, must-follow sync rules and working style.
orloj CLAUDE.md
Instructions for OrlojHQ/orloj, covering orloj claude instructions, critical repository rules and implementation notes.
agentfootprint AGENTS.md
Instructions for footprintjs/agentfootprint, covering agentfootprint — agent instructions (openai codex), core thesis, the mental model — three slots, six flavors, mock-first development (recommended workflow) and public api.
permit0 CLAUDE.md
Instructions for permit0-ai/permit0, covering claude.md, commands, build, test (ci uses nextest) and run a single test.
agentfootprint CLAUDE.md
Instructions for footprintjs/agentfootprint, covering agentfootprint — feature-work map, before you design it: it may already exist, module map, core state & flow and extension points.
filegrc AGENTS.md
Instructions for Alignbase/filegrc, covering filegrc repository instructions, purpose, agent-facing product surface, product principles and standards alignment.