srt-title-generator
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from external SRT files which could theoretically contain malicious instructions designed to manipulate the agent's output.
- Ingestion points: The skill uses
catto read the transcript content from user-provided SRT files into the agent's context. - Boundary markers: The prompt framework uses a
<scratchpad>for reasoning but does not employ explicit delimiters or system instructions to ignore potentially embedded commands within the SRT content. - Capability inventory: The skill is restricted to reading local files and generating text; it lacks network access, file writing capabilities, or the ability to execute code, which significantly mitigates the risk of an indirect injection.
- Sanitization: The skill mentions stripping timestamps and formatting, but does not perform security-focused sanitization of the dialogue text.
Audit Metadata