organize-skills

Pass

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill employs various shell commands such as mv, git mv, ls, readlink, find, and grep to inventory and reorganize skill repositories. These operations modify the file system and version control history based on the existing structure of the user's skill hubs and project directories.\n- [PROMPT_INJECTION]: The skill exposes an Indirect Prompt Injection surface by ingesting and acting upon content found in third-party SKILL.md files and referenced project data.\n
  • Ingestion points: The agent reads and processes all files within designated skill hubs and shared libraries, including SKILL.md frontmatter and repository source code during classification and cross-reference fixing.\n
  • Boundary markers: There are no explicit delimiters or instructions to ignore embedded instructions when the agent parses or greps through the content of the skills being audited.\n
  • Capability inventory: The skill possesses the ability to move directories, manipulate symlinks, commit changes to git repositories, and modify the text content of files across the user's library and memory directories.\n
  • Sanitization: No sanitization or validation logic is applied to the data retrieved from external skill files before it is used to drive the reorganization process, potentially allowing malicious content in an audited skill to influence the agent's actions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 14, 2026, 09:00 AM
Security Audit — agent-trust-hub — organize-skills