subtitle-correction
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- [NO_CODE] (SAFE): The provided files (srt-format.md and terminology.md) are purely informational Markdown documents and do not include any scripts, binaries, or automated tasks.
- [PROMPT_INJECTION] (LOW): The skill possesses a surface for indirect prompt injection (Category 8) because it is designed to process untrusted subtitle text. 1) Ingestion points: Subtitle text lines within SRT files processed according to srt-format.md. 2) Boundary markers: The instructions use the rigid SRT structure (Sequence Numbers and Timestamps) as natural delimiters for data. 3) Capability inventory: No dangerous capabilities, such as network access or command execution, are available in the skill. 4) Sanitization: The skill provides explicit 'DO' and 'DON'T' rules that restrict the agent's actions to specific text corrections, reducing the impact of potential injections.
Audit Metadata