skills/sergio-bershadsky/ai/version/Gen Agent Trust Hub

version

Warn

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • COMMAND_EXECUTION (MEDIUM): The skill performs shell interpolation of the plugin name and version variables in commands such as git commit -m "chore(<name>): bump version to <new-version>" and cat plugins/<name>/.claude-plugin/plugin.json. Malicious input (e.g., ;, &, or $(...)) in the plugin name or version field could lead to arbitrary command execution or path traversal beyond the intended directory.\n- INDIRECT_PROMPT_INJECTION (LOW): The skill ingests data from local configuration files which could be manipulated to influence agent behavior.\n
  • Ingestion points: File content from plugin.json and marketplace.json is read in Step 2 and Step 5.\n
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the processing logic.\n
  • Capability inventory: The skill has access to directory listing (ls), file reading (cat), and version control operations (git add, git commit, git tag).\n
  • Sanitization: No validation or escaping is performed on the extracted version string or the user-provided plugin name.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 17, 2026, 06:27 PM