push-release

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill takes user input to determine the version increment level (e.g., patch, minor, major) and interpolates it directly into a shell command: make bump_version BUMP=<user_input>. If the agent does not validate this input against a strict allow-list, an attacker could provide a malicious string containing command separators (like ; or &&) to execute arbitrary commands on the system.
  • Ingestion points: The BUMP level provided by the user (Step 2).
  • Boundary markers: None present in the skill instructions to guide the agent on sanitizing the input.
  • Capability inventory: The skill uses git and make to modify files and interact with remote repositories (Step 1, 3, and 4).
  • Sanitization: No explicit sanitization or validation logic is defined in the instructions to ensure the BUMP variable only contains safe values.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 10:53 AM