bilibili-chapter-generator

Fail

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: HIGHCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION] (HIGH): Potential shell command injection via unvalidated file paths in SKILL.md.
  • The workflow in Step 2 executes python3 ~/.claude/skills/srt-to-structured-data/scripts/parse_srt.py "<srt_file_path>" --stats.
  • Because the <srt_file_path> is interpolated directly into the command string, a malicious user can provide a path like "; rm -rf /; " to execute arbitrary shell commands.
  • [PROMPT_INJECTION] (LOW): Risk of indirect prompt injection through untrusted subtitle data ingestion.
  • Ingestion points: Reads external .srt files provided by the user (SKILL.md Step 1).
  • Boundary markers: Absent; the skill does not wrap content in delimiters or warn the agent about embedded instructions.
  • Capability inventory: Subprocess execution via python3 (SKILL.md Step 2).
  • Sanitization: None; the skill lacks validation for the input path or the contents of the SRT file before processing.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 17, 2026, 06:06 PM