report
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (LOW): The skill executes
mkdir -pto manage report directories. While common for utility skills, this involves local command execution.\n- [PROMPT_INJECTION] (LOW): The skill possesses an indirect prompt injection surface as it interpolates untrusted investigation data into reports.\n - Ingestion points: External research, debugging, and audit data processed in
SKILL.md.\n - Boundary markers: Absent; the report template does not provide delimiters to distinguish summarized data from instructions.\n
- Capability inventory:
SKILL.mduses subprocess calls (mkdir) and file-system write operations.\n - Sanitization: Absent; there are no instructions to sanitize input data or the
topicvariable for command injection or markdown bypass patterns.
Audit Metadata