red-team-agent-orchestration: Command for Claude Code

.claude/commands/attack-paths.md

attack-paths is a command for Claude Code from Contoso-State/red-team-agent-orchestration. It costs 23 tokens per session (442 once invoked), scanned A, original, MIT.

An Azure security analysis command that combines separate findings into multi-step attack paths. Azure is Microsoft’s cloud platform; an attack path shows how an attacker could move from an entry point to a more serious outcome.

In plain words
What is it for?
Use it after Azure assessment agents have produced raw findings. It traces routes involving privilege escalation, movement between environments, identity abuse, secret stores, data, and compute.
Why use it?
It turns isolated permission or configuration findings into realistic chains, making the likely impact and escalation route clearer.

Command for Claude Code

Written for Claude Code: installed under .claude/.

This is Contoso-State/red-team-agent-orchestration's own configuration. It tells Claude Code how to work on red-team-agent-orchestration 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 red-team-agent-orchestration configures →

Reuse

Borrowing it

Nothing to install: this file belongs to Contoso-State/red-team-agent-orchestration. 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/Contoso-State/red-team-agent-orchestration/main/.claude/commands/attack-paths.md
Clone the repo
git clone --depth 1 https://github.com/Contoso-State/red-team-agent-orchestration

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 attack-paths

README.md
[![agentmods](https://agentmods.dev/badge/commands/contoso-state/red-team-agent-orchestration/attack-paths.svg)](https://agentmods.dev/commands/contoso-state/red-team-agent-orchestration/attack-paths)
Your own site
<a href="https://agentmods.dev/commands/contoso-state/red-team-agent-orchestration/attack-paths"><img src="https://agentmods.dev/badge/commands/contoso-state/red-team-agent-orchestration/attack-paths.svg" alt="Measured on agentmods" height="20"></a>
Per session 23 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 442 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 $0.00023 $0.00442
Opus 5 $0.00012 $0.00221
Sonnet 5 $0.00005 $0.00088
Haiku 4.5 $0.00002 $0.00044

Measured 7d ago against content hash 103907e2bf08, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

attack-paths 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 7d 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/commands/attack-paths.md · 32 lines

What it actually says

/attack-paths — Attack Path Analysis

You are acting as the Authorization & Attack Path Agent (agents/authorization-attack-path/system-prompt.md). Correlate the raw findings into real, walkable attack chains.

Preconditions

  • Domain agents have run; engagements/<session>/findings/raw/*.jsonl is populated.

Steps

  1. Load all findings from engagements/<session>/findings/raw/*.jsonl and the resource inventory.
  2. Build the privilege graph: principals × role assignments × scopes × managed identities × resources.
  3. Trace chains. Look for the classic Azure compromise patterns:
    • Public/internet-facing entry point → managed identity → secret store → privileged data or compute
    • Low-privilege principal holding an escalation primitive (roleAssignments/write, runCommand/action, listClusterAdminCredential/action, etc.) → self-elevation
    • Cross-environment movement via VNet peering, shared identities, or shared Key Vaults
    • Contributor/Owner on a scope → modify workload → steal its identity token
  4. Score by end state. A chain's severity reflects the final impact, not the weakest individual step.
  5. Emit chain findings to engagements/<session>/findings/raw/authorization-attack-path.jsonl with attack_path populated and ID prefix AZ-PATH-.

Output

  • Correlated attack-path findings, each with an ordered attack_path
  • A ranked list of the most dangerous chains with their entry point and end state

Read-only analysis only. Describe the paths; do not execute them unless controlled-validation mode explicitly permits and engagement.yaml allows the action.

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. 7d ago First seen · 32 lines · 23 tokens per session scan A 103907e2bf08

Subscribe to this mod's changes

attack-paths is a command published in the GitHub repository Contoso-State/red-team-agent-orchestration (6 stars, last pushed 3d ago), licensed MIT. It adds 23 tokens to every session and 442 once invoked, about $0.0001 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.