skills/poteto/plugins/thermos/Gen Agent Trust Hub

thermos

Pass

Audited by Gen Agent Trust Hub on Aug 19, 2026

Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
  • [SAFE]: Analysis of the skill instructions confirms they are limited to workflow orchestration and do not involve any malicious activities such as data exfiltration or credential theft.
  • [NO_CODE]: The skill does not contain any executable scripts, binaries, or configuration files that could hide malicious logic. It relies entirely on standard platform subagents.
  • [PROMPT_INJECTION]: The skill's primary function is to analyze untrusted code diffs and PR descriptions, which serves as a surface for indirect prompt injection.
  • Ingestion points: Pull request diffs, branch file context, and user-provided request strings (SKILL.md).
  • Boundary markers: None specified in these instructions; relies on the security posture of the called subagents.
  • Capability inventory: Orchestrates subagent execution and synthesizes their textual output.
  • Sanitization: No sanitization steps for input diffs are described in the workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 19, 2026, 03:44 PM
Security Audit — agent-trust-hub — thermos