lyrics-overlay
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines an attack surface for indirect prompt injection by ingesting and processing external transcript data.
- Ingestion points: The
parseSRTandparseElevenLabsTranscriptfunctions inremotion-templates/utils/lyricsParser.tsingest text content from external SRT or JSON files. - Boundary markers: No explicit delimiters or instructions are provided to the agent to treat transcript content as untrusted data or to ignore embedded instructions.
- Capability inventory: The skill's components are designed for visual rendering within the Remotion environment. No dangerous capabilities such as network exfiltration, arbitrary command execution, or file-writing operations are present in the provided scripts.
- Sanitization: The text is rendered via standard React components, which provides inherent protection against XSS but does not filter for Natural Language instructions aimed at the agent.
- [SAFE]: The skill uses standard development practices for the Remotion framework, including well-known React hooks and animation utilities. References to external tools like
npx remotionand thetranscribeskill are consistent with the primary purpose of video production.
Audit Metadata