edit-plugin

Pass

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes local shell scripts and standard command-line tools including git, find, and grep to identify project changes and automate version metadata updates.\n- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by ingesting untrusted repository metadata.\n
  • Ingestion points: File names and directory names within the plugin folders, as well as Git tags retrieved via git describe in the detect-changes.sh script.\n
  • Boundary markers: The shell script output is formatted as YAML without using delimiters or instructions to ignore embedded commands within the extracted data.\n
  • Capability inventory: The agent uses the Edit tool to perform automated file modifications and version updates based on the processed metadata.\n
  • Sanitization: No validation or escaping is applied to the extracted names or tags before they are incorporated into the agent's decision-making logic, allowing potentially malicious metadata to influence the versioning workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 2, 2026, 06:18 AM
Security Audit — agent-trust-hub — edit-plugin