devil

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from external sources, establishing a surface for indirect prompt injection attacks.
  • Ingestion points: Document content is retrieved via file paths, URLs, or direct text input as defined in the 'Getting the document' section of SKILL.md.
  • Boundary markers: The skill lacks explicit instructions to use delimiters or ignore embedded commands within the analyzed documents, which may lead the agent to follow instructions contained within a spec.
  • Capability inventory: The agent is directed to use available file-reading and web-fetching tools (e.g., Confluence MCP) to retrieve content for analysis.
  • Sanitization: There are no instructions for sanitizing, validating, or escaping the retrieved document content before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 02:12 PM
Security Audit — agent-trust-hub — devil