pese
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill refers to official documentation and software from Tailscale and its Tailcat project on GitHub to establish secure communication channels.
- [COMMAND_EXECUTION]: The skill uses local command-line interfaces for Tailscale and Tailcat to configure private transport routes and verify resource availability.
- [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface where it processes local data which could potentially contain malicious instructions; however, it mitigates this risk through strict boundary enforcement and resource validation.
- Ingestion points: Local files, directories, and web services specified by the user in SKILL.md.
- Boundary markers: Includes explicit instructions to reject home directories, credential files, and unresolved symlinks.
- Capability inventory: Performs subprocess calls to networking tools and manages ephemeral local listeners as described in SKILL.md and references/tailcat.md.
- Sanitization: Requires pinning of exact resources and the use of allowlists for any required companion assets.
Audit Metadata