distill
Pass
Audited by Gen Agent Trust Hub on Mar 25, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill is defined entirely within a markdown file and contains no executable scripts, source code, or binaries.
- [SAFE]: No security threats such as credential harvesting, data exfiltration, or unauthorized command execution were detected.
- [SAFE]: The skill prompts the agent to gather context from the 'current thread or codebase'. While this constitutes a data ingestion surface, the skill lacks any tools or capabilities (e.g., file-write, network access, or subprocess execution) that could be leveraged for an attack.
- Ingestion points: thread context and codebase; Boundary markers: absent; Capability inventory: none; Sanitization: absent.
- [SAFE]: Reference to a 'frontend-design' skill is an internal platform configuration for modularity and does not involve external downloads or unverifiable dependencies.
- [SAFE]: No obfuscation, hidden instructions, or prompt injection patterns were found in the metadata or instructional body.
Audit Metadata