supsub-unread

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFE
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill provides an installation command that pipes a remote script to bash: curl -fsSL https://raw.githubusercontent.com/SupSub-AI/supsub-cli/master/scripts/install.sh | bash. This script originates from the author's official GitHub organization and is used to install the necessary CLI environment.
  • [EXTERNAL_DOWNLOADS]: The skill instructs users to install the @supsub/cli package via npm or pnpm, which is the official package provided by the vendor for this skill's functionality.
  • [COMMAND_EXECUTION]: The skill interacts with the local file system and network by executing supsub CLI commands such as supsub unread, supsub sub contents, and supsub focus mark-read to manage subscription data.
  • [DATA_EXPOSURE]: The skill processes content from external subscription feeds, representing an indirect prompt injection surface.
  • Ingestion points: Subscription content retrieved via supsub sub contents and supsub focus contents (SKILL.md).
  • Boundary markers: The skill recommends using -o json for parsing, but does not specify delimiters for isolating untrusted text within the prompt.
  • Capability inventory: The skill executes CLI commands to read and modify subscription states.
  • Sanitization: No explicit sanitization or validation of the content from the subscription feeds is mentioned in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 01:15 AM
Security Audit — agent-trust-hub — supsub-unread