open-knowledge-format

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes OKF bundles consisting of user-supplied Markdown files. This introduces an indirect prompt injection surface where instructions embedded in files could influence agent behavior. (1) Ingestion points: Markdown files are read from local directories by the agent and scripts. (2) Boundary markers: The skill relies on YAML frontmatter but lacks explicit sanitization for Markdown bodies. (3) Capability inventory: The skill allows for file system access and execution of local Ruby scripts. (4) Sanitization: No sanitization of external content is implemented.
  • [COMMAND_EXECUTION]: The skill provides and instructs the execution of Ruby scripts for bundle validation and package digest generation. These scripts operate on local file paths.
  • [EXTERNAL_DOWNLOADS]: The skill fetches and references documentation from the official Google Cloud Platform GitHub repository for format conformance. This is a neutral reference to a well-known service.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 03:30 PM
Security Audit — agent-trust-hub — open-knowledge-format