hugo-expert

Pass

Audited by Gen Agent Trust Hub on Aug 7, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests data from untrusted local files (Hugo configurations, CI scripts, and directory structures) to provide advice. There are no explicit boundary markers or instructions to the agent to ignore potentially malicious commands within these files.
  • Ingestion points: The detect_hugo.py script reads hugo.toml, netlify.toml, and files within the .github/workflows directory.
  • Boundary markers: Absent in both the Python script and the main SKILL.md instructions.
  • Capability inventory: The skill enables the agent to run shell commands and potentially modify repository files.
  • Sanitization: No sanitization is performed on the data read from the repository files.
  • [COMMAND_EXECUTION]: The skill executes a local Python script (detect_hugo.py) and standard Hugo commands. The Python script is verified to be safe, using only built-in libraries for file analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 7, 2026, 03:08 PM
Security Audit — agent-trust-hub — hugo-expert