microsoft-security-agent-toolkit: Instructions file for Claude Code

CLAUDE.md

microsoft-security-agent-toolkit CLAUDE.md is an instructions file for Claude Code from vinayaklatthe/microsoft-security-agent-toolkit. It costs 549 tokens per session, scanned A, original, MIT.

Repository-specific instructions for working on Microsoft Security Agent Toolkit, a codebase that provides tools and templates for AI agents working with Microsoft security systems. They describe its context, rules, conventions, and unfinished areas.

In plain words
What is it for?
Use them when modifying this repository, choosing the right tool identifiers, following its commit conventions, writing tests or demos with synthetic data, and checking project decisions in HANDOFF.md.
Why use it?
They give coding agents the project facts and safety rules they need before changing the repository, such as protecting tenant secrets and requiring confirmation for changes to live systems.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: mentions Claude Code.

This is vinayaklatthe/microsoft-security-agent-toolkit's own configuration. It tells Claude Code how to work on microsoft-security-agent-toolkit itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything microsoft-security-agent-toolkit configures →

Reuse

Borrowing it

Nothing to install: this file belongs to vinayaklatthe/microsoft-security-agent-toolkit. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/vinayaklatthe/microsoft-security-agent-toolkit/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/vinayaklatthe/microsoft-security-agent-toolkit

Made for: Claude Code.

Wrote 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.

agentmods badge for microsoft-security-agent-toolkit CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/vinayaklatthe/microsoft-security-agent-toolkit/claude-md.svg)](https://agentmods.dev/instructions/vinayaklatthe/microsoft-security-agent-toolkit/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/vinayaklatthe/microsoft-security-agent-toolkit/claude-md"><img src="https://agentmods.dev/badge/instructions/vinayaklatthe/microsoft-security-agent-toolkit/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 549 This file is loaded in full into every session.
When invoked 549 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00549 $0.00549
Opus 5 $0.00275 $0.00275
Sonnet 5 $0.00110 $0.00110
Haiku 4.5 $0.00055 $0.00055

Measured 6d ago against content hash 181a9bfe6285, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

microsoft-security-agent-toolkit CLAUDE.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 6d 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.

CLAUDE.md · 35 lines

How it starts

The opening of the file, as written. The whole thing — 35 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Claude Code instructions for microsoft-security-agent-toolkit

Always read HANDOFF.md at the repo root before making any change. It is the source of truth for design decisions, conventions, tool surface, and open work prioritization.

Quick context

  • Repo role: action layer (MCP servers, KQL snippets, Logic Apps templates, demos) for AI agents working on Microsoft Security.
  • Pairs with: microsoft-security-skills (the knowledge layer, 88 skills, v0.2.0).
  • Status: preview — scaffold complete, MCP implementations pending.

Hard rules (do not violate)

  1. Never log or commit a tenant secret. .env stays gitignored.
  2. Every write tool (mutates tenant state) must take confirm: true and use the separate write-app registration. See docs/permissions-required.md.
  3. Skills only declare tools_required — they never bundle this toolkit. The host routes.
  4. No offensive (red-team) tooling — defensive and governance only.
  5. No live-tenant calls in tests, demos, or samples — synthetic data only.

Conventions

  • Conventional-commit-ish (scaffold:, feat(sentinel-mcp):, docs:, kql:). Always add Co-authored-by: Copilot <[email protected]> when AI-assisted.
  • Branch off main. PR into main. One server / one snippet pack / one demo per PR.
  • KQL files use the header documented in kql-snippets/README.md.

Tool IDs (do not rename without cross-repo update)

sentinel.run_kql, sentinel.list_incidents, sentinel.get_incident, sentinel.update_incident (write) · defender_xdr.list_alerts, defender_xdr.get_incident, defender_xdr.run_advanced_hunting, defender_xdr.update_incident (write), defender_xdr.isolate_device (write) · graph.signins, graph.risky_users, graph.directory_audit, graph.revoke_sessions (write), graph.disable_user (write) · purview_audit.search, purview_audit.export

When stuck

Pick the topmost unchecked item from "Open work" in HANDOFF.md. Phase 1 (sentinel-mcp end-to-end implementation) is the current priority.

Read the full file on GitHub · 35 lines

Changes

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.

  1. 6d ago First seen · 35 lines · 549 tokens per session scan A 181a9bfe6285

Subscribe to this mod's changes

microsoft-security-agent-toolkit CLAUDE.md is an instructions file published in the GitHub repository vinayaklatthe/microsoft-security-agent-toolkit (2 stars, last pushed 2mo ago), licensed MIT. It adds 549 tokens to every session, about $0.0027 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 2026-08-31.