bilibili-chapter-generator
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script to parse subtitle files.
- Evidence:
python3 ~/.claude/skills/srt-to-structured-data/scripts/parse_srt.py "<srt_file_path>" --statsinSKILL.md. - [INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes untrusted content from external SRT subtitle files.
- Ingestion points: The output of the
parse_srt.pyscript which contains text from user-provided SRT files. - Boundary markers: Absent. There are no explicit instructions to the agent to treat the subtitle text as untrusted data or to ignore potential instructions embedded within the subtitles.
- Capability inventory: The skill performs local command execution of a specific parsing script but does not demonstrate broader network or file-writing capabilities.
- Sanitization: Absent. The skill does not mention filtering or sanitizing the subtitle content before analysis.
Audit Metadata