sfd-media-critic

Pass

Audited by Gen Agent Trust Hub on Apr 10, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill defines a standard interface for media analysis and does not contain any suspicious commands, network exfiltration patterns, or obfuscated content.
  • [NO_CODE]: The provided files consist of documentation and configuration instructions only; no executable scripts (Python, JavaScript, shell) are included in the skill.
  • [SAFE]: Potential indirect prompt injection surface related to media content analysis.
  • Ingestion points: Processes external media via mediaUrl and description text in SKILL.md.
  • Boundary markers: Prompts utilize placeholders but do not define explicit escape sequences or instructions to ignore embedded content.
  • Capability inventory: Limited to calling the /v1/watch tool; no file system access, shell command execution, or network writing capabilities are present across the files.
  • Sanitization: Standard tool-based processing; the agent's internal safety filters act as the primary defense for generated output.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 10, 2026, 01:30 AM