ldm-ai-to-human-zh
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the AI agent to execute a local Python script,
scripts/hanyihan_qc.py, to perform text analysis and quality checks on user-provided files. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted text data provided by users to identify AI-style patterns.
- Ingestion points: The
hanyihan_qc.pyscript reads the content of markdown files provided as command-line arguments. - Boundary markers: The instructions explicitly require the agent to generate a report and obtain user confirmation before performing any modifications, which serves as a human-in-the-loop safety boundary.
- Capability inventory: The skill includes file reading, execution of its own bundled Python script, and creation of new versioned files (e.g.,
filename_V2.md). - Sanitization: The Python script uses regular expressions for pattern matching and text analysis; it does not execute the content of the analyzed files.
Audit Metadata