skills/freestylefly/author-methodology-analysis-skill/author-methodology-analysis/Gen Agent Trust Hub
author-methodology-analysis
Pass
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes internal Python scripts (analyze_corpus.py, generate_dashboard.py, and validate_outputs.py) to process data and generate reports. These scripts use standard libraries and operate on local data without performing dangerous or unauthorized actions.
- [DATA_EXFILTRATION]: The skill offers an integration to sync results to Lark (Feishu), a well-known enterprise collaboration service. This feature includes a proactive safety check in the corpus analysis script that detects sensitive PII (Chinese identity numbers, phone numbers, emails) and sets a 'requires_confirmation' flag to ensure user review before proceeding with network synchronization.
- [PROMPT_INJECTION]: As the skill is designed to ingest and analyze arbitrary external documents, it has an inherent vulnerability to indirect prompt injection. 1. Ingestion points: Document files (Markdown, TXT, DOCX, PDF) in the provided input directory. 2. Boundary markers: None explicitly defined in the instructions for the interpretive analysis phase. 3. Capability inventory: Local file system access, script execution, and network synchronization. 4. Sanitization: Content is cleaned via markdown stripping in the preprocessing script.
Audit Metadata