skills/lgbarn/skills/compass-ingest/Gen Agent Trust Hub

compass-ingest

Pass

Audited by Gen Agent Trust Hub on Jun 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses standard shell commands including ls, mv, and git mv to locate, move, and rename files within the local filesystem. These operations are restricted to the user's Downloads and specific project directories.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) because it reads and processes untrusted content from files located in the ~/Downloads folder to determine file naming and next steps.
  • Ingestion points: Content is read from local files matching the pattern ~/Downloads/compass_artifact_* in SKILL.md.
  • Boundary markers: None. The instructions do not specify the use of delimiters or 'ignore' instructions when reading the artifact content.
  • Capability inventory: The skill has the ability to move/rename files and trigger subsequent tasks such as metal-sweep based on the content read.
  • Sanitization: There is no explicit validation or sanitization mentioned for the content-derived 'topic' string used in the final filename.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 30, 2026, 03:29 PM
Security Audit — agent-trust-hub — compass-ingest