sillytavern-media-live2d-runtime
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions provide shell commands for executing included Node.js scripts (
validate-media-manifest.mjs,check-media-bindings.mjs,plan-media-preload.mjs) to perform validation and planning tasks. These scripts are part of the skill's distribution and operate on local data. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied JSON files (manifests and bindings). While this introduces an input surface, the risk is mitigated by strong validation logic within the provided tools.
- Ingestion points: The skill reads
media-manifest.json,media-bindings.json, andpreload-budget.jsonvia dedicated validation scripts. - Boundary markers: Not explicitly defined for tool output reporting.
- Capability inventory: Includes file reading capabilities (restricted to a root directory) and cryptographic hash verification.
- Sanitization: Included scripts implement strict schema validation, type checking, and path traversal prevention using
path.resolveand prefix checks.
Audit Metadata