identity-provider-attack

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external identity artifacts such as JWT tokens and SAML assertions, which creates a vulnerability surface for indirect prompt injection.
  • Ingestion points: The skill processes external identity data blocks as demonstrated in payloads.md and test-cases.md (e.g., JWT strings, XML assertions).
  • Boundary markers: Absent. The instructions do not define clear delimiters or specify that the agent should ignore instructions embedded within the processed identity data.
  • Capability inventory: The skill utilizes Bash, Write, Edit, and WebFetch tools, providing a broad capability set that could be abused if malicious instructions in an ingested token were executed.
  • Sanitization: Absent. There is no evidence of validation or sanitization routines for the tokens before they are processed by the agent's logic or passed to shell tools.
  • [COMMAND_EXECUTION]: The skill provides instructions for executing various third-party security tools and scripts via the command line. While these are appropriate for offensive security testing, they involve executing external code.
  • Evidence: References to python3 jwt_tool.py, o365creeper.py, SAMLExtractor.py, and tools like mitm6, AADInternals, and ROADtools in SKILL.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 04:36 PM