skills/openclaw/skills/bat-cat/Gen Agent Trust Hub

bat-cat

Pass

Audited by Gen Agent Trust Hub on Mar 7, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious code, credential theft, or unauthorized access patterns were identified. The skill acts exclusively as a documentation and installation wrapper for a well-known open-source tool.
  • [EXTERNAL_DOWNLOADS]: The skill's metadata facilitates the installation of the 'bat' binary through official repositories for Homebrew and APT. These are well-known and trusted package management services.
  • [PROMPT_INJECTION]: The skill facilitates the ingestion of untrusted data from local files, creating a surface for indirect prompt injection if the agent interprets the tool's output as instructions.
  • Ingestion points: Commands such as bat README.md and bat file.json in SKILL.md demonstrate the reading of external file content.
  • Boundary markers: No specific delimiters or instructions to ignore embedded content are included in the usage examples.
  • Capability inventory: The skill provides the ability to execute the bat binary to read and display file content.
  • Sanitization: No sanitization is performed by the tool; security relies on the agent's internal handling of ingested text.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 7, 2026, 07:32 PM