supsub-sub

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructions and commands are consistent with its stated purpose of managing subscriptions via the supsub CLI.
  • [EXTERNAL_DOWNLOADS]: The skill references the installation of the @supsub/cli package. This is a vendor-owned resource (SupSub-AI) and is a standard prerequisite for the skill.
  • [COMMAND_EXECUTION]: The skill uses local shell commands (supsub sub list, supsub sub add, etc.) to interface with the subscription service. These commands are restricted to the vendor's own tooling and include guidance on handling browser-based authentication.
  • [PROMPT_INJECTION]: The skill contains instructions for the agent to correctly resolve ambiguity in user requests (e.g., the letter 'X' referring to a placeholder versus the Twitter platform) and to prioritize subscription browsing over keyword search when appropriate.
  • [PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection as it processes external content (titles and summaries) from subscriptions.
  • Ingestion points: supsub sub contents output (article metadata) in SKILL.md.
  • Boundary markers: None explicitly mentioned in the skill text.
  • Capability inventory: The skill is limited to listing and displaying information; it does not perform high-privilege file writes or arbitrary network operations based on the ingested data.
  • Sanitization: Not explicitly implemented in the skill instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 01:15 AM
Security Audit — agent-trust-hub — supsub-sub