blog-post

Pass

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: SAFE
Full Analysis
  • [DYNAMIC_CONTEXT_INJECTION]: The skill employs the !command`` syntax in its Context section to execute find and git commands (e.g., git remote -v, git branch --show-current) when the skill is loaded. These commands are used to automatically populate repository information into the agent's context and do not perform unauthorized network operations or access sensitive system files.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests untrusted data from git log (commit messages) and user-supplied arguments to build blog post templates.
  • Ingestion points: Reads commit messages via git log and project names via git remote.
  • Boundary markers: Absent. The skill does not explicitly warn the model to ignore instructions found within git history.
  • Capability inventory: The skill is limited to reading/writing files and executing specific, constrained bash commands (hugo, date).
  • Sanitization: No specific sanitization or escaping of git output is performed before it is presented to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 21, 2026, 01:17 AM