ljg-skill-map
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs glob-based scanning of the local directory
~/.claude/skills/ljg-*/to readSKILL.mdfiles. This access is restricted to the specific vendor namespace ('ljg-*') and only processes non-sensitive metadata (name and description) for mapping purposes.\n- [COMMAND_EXECUTION]: Employs the systemdatecommand for timestamp generation in documentation templates. These are standard operations and do not involve the execution of untrusted code or unsanitized user input.\n- [PROMPT_INJECTION]: Evaluated for Indirect Prompt Injection (Category 8): (1) Ingestion points: reads frontmatter metadata from local files. (2) Boundary markers: absent. (3) Capability inventory: local file system read (glob) and terminal output only. (4) Sanitization: truncates descriptions to 100 characters. The risk is negligible as the agent is instructed to produce a strictly formatted ASCII topology map, which limits the potential for external strings to hijack the agent's logic.
Audit Metadata