post-exploiting-microsoft-graph-with-graphrunner

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill repeatedly instructs passing access/refresh tokens into commands (e.g., -AccessToken/-RefreshToken/-Tokens) which requires the LLM/agent to handle and potentially emit secret token values verbatim, creating a high exfiltration risk.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). The GitHub links point to a public PowerShell post‑exploitation repository (GraphRunner) and an associated offensive tool (ROADtools) that deliver scripts/executables capable of abusing Microsoft Graph and establishing persistence, so they are legitimate distribution sources for potentially malicious code.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The repository and accompanying scripts are explicit post-exploitation tools for Microsoft 365/Entra that obtain/handle access tokens, deploy malicious OAuth apps and inbox forwarding, manipulate groups for persistence/escalation, and search/download mailbox/Drive data — behavior clearly intended for credential theft, persistence, and data exfiltration.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The SKILL.md explicitly instructs git clone https://github.com/dafthack/GraphRunner.git followed by Import-Module .\GraphRunner.ps1, which fetches remote PowerShell code and then imports/executes it at runtime (https://github.com/dafthack/GraphRunner.git).