compile-knowledge

Pass

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill defines a structured procedure for knowledge management and does not contain any malicious patterns or instructions to bypass safety guidelines.- [COMMAND_EXECUTION]: The skill mentions a command line tool 5dive memory add. This is identified as a vendor-specific resource belonging to the author '5dive-ai', used for its stated purpose of updating the agent's memory files.- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the storage of external data which constitutes a potential attack surface for indirect prompt injection.
  • Ingestion points: The agent is instructed to store research, intelligence, digests, and learned facts into its memory files.
  • Boundary markers: The instructions recommend using markdown frontmatter (name, description, type) and atomic file structures to organize the data.
  • Capability inventory: The skill uses the 5dive memory add command to write to local storage within the .claude/ and wiki/ directories.
  • Sanitization: The skill does not provide specific sanitization or escaping instructions for the content being stored, relying on default agent behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 28, 2026, 12:33 AM
Security Audit — agent-trust-hub — compile-knowledge