
Auditing Entra ID with AADInternals
Runs AADInternals to test Microsoft Entra ID tenant exposure and AD FS Golden SAML resilience during authorized assessments.
- Repository
- mukul975/Anthropic-Cybersecurity-Skills
- Stars
- 35,430
- Licence
- Apache-2.0
- Files
- 4 files
What it does
Once installed, the agent can drive AADInternals, a PowerShell toolkit for Microsoft Entra ID (formerly Azure AD) and AD FS, to run a structured identity security assessment. It walks through unauthenticated reconnaissance of a domain's tenant and federation settings, optional user enumeration, token acquisition for Microsoft APIs, directory enumeration with a valid token, and checks of whether federation signing certificates would allow Golden SAML forgery. It then helps document findings and hardening recommendations. It is built for authorized red-team and defensive validation work, not general use.
Great for
- Security consultants running an authorized Entra ID or Microsoft 365 red-team engagement
- Blue teams validating that AD FS token-signing certificates are protected against Golden SAML
- Detection engineers who need realistic AADInternals telemetry to build alerts
Skip it if
- Testing any tenant you do not own or lack written authorization for, since the toolkit can forge SAML tokens and install persistent federation backdoors (legal and safety risk)
- Casual or beginner use, because it assumes familiarity with SAML, OAuth tokens and Azure AD Connect and requires Global Administrator rights for the backdoor tests
Try asking
- “Run outsider reconnaissance on target.com to list its tenant ID, federation type and SSO status (authorized engagement)”
- “Check whether a tenant's AD FS token-signing certificate is protected against Golden SAML and write up the evidence”
- “Enumerate which domains are federated and recommend hardening for the federation configuration”
- “Map which Microsoft APIs a low-privileged token can reach and document the exposure”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 4 original files, the Apache-2.0 licence and an attribution note.
auditing-entra-id-with-aadinternals.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip auditing-entra-id-with-aadinternals.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip auditing-entra-id-with-aadinternals.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Confirm written scope and rules of engagement before any command runs; keep federation-backdoor tests inside the approved window
- Start with the no-credential outsider recon steps, which only query public Microsoft endpoints, before moving to authenticated steps
- Ask the agent to log every cmdlet and output for the evidence package and to recommend remediation rather than exploit further
What's inside
- references/api-reference.md
- references/standards.md
- scripts/agent.py
- SKILL.md
Redistributed unmodified from mukul975/Anthropic-Cybersecurity-Skills under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.