moviepilot-update
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill interacts with external data (version information and update status) which could theoretically serve as an injection vector, though the risk is mitigated by the restricted API structure.
- Ingestion points: The
system.versions,system.update.status, andsystem.update.checkoperations inSKILL.mdingest data from the MoviePilot service. - Boundary markers: No explicit delimiters are defined in the instructions for data processing.
- Capability inventory: The skill has the capability to restart the service (
system.restart) and install updates (system.update.install), which are privileged actions. - Sanitization: The skill relies on the platform's tool-calling architecture and explicitly instructs the agent to follow host-side confirmation policies.
- [SAFE]: The skill follows security best practices by explicitly forbidding the agent from requesting or passing sensitive data like API tokens, custom URLs, or raw shell commands.
- [SAFE]: All destructive or state-changing operations (restarts and installations) are documented to require explicit human-in-the-loop confirmation before execution.
Audit Metadata