doc-source-reconciliation

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill takes an untrusted document as input and extracts 'verifiable facts' to check against a repository. This creates a surface for indirect prompt injection where malicious instructions embedded in the document could influence the agent's behavior.
  • Ingestion points: The input document provided to the agent as described in SKILL.md.
  • Boundary markers: Absent. There are no instructions to wrap the untrusted document in specific delimiters or to treat its contents strictly as data.
  • Capability inventory: The agent is instructed to perform file system reads and searches within a provided repository path to verify claims.
  • Sanitization: Absent. There is no mention of sanitizing or validating the claims extracted from the document before they are used to guide file system operations.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 10:46 AM