md-to-feishu
Fail
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: Provides instructions to download and execute the installation script for the 'uv' tool from astral.sh using a pipe to shell (curl | sh).
- [COMMAND_EXECUTION]: Invokes local CLI tools such as 'uvx', 'feishu-docx', and 'python3' to perform environment checks and document conversion tasks.
- [EXTERNAL_DOWNLOADS]: Fetches rendered Mermaid diagrams as PNG files from the mermaid.ink API and installs software packages via pip and uvx.
- [DYNAMIC_EXECUTION]: Includes a Python script template that dynamically constructs URLs and performs network requests using 'urllib.request' to render diagrams at runtime.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided Markdown files to extract titles and render diagrams, which could contain malicious instructions.
- Ingestion points: Reads content from user-specified local Markdown files in Step 1 and Step 3.
- Boundary markers: None identified; instructions do not include delimiters or specific guidance to ignore instructions within the Markdown content.
- Capability inventory: The skill has capabilities for file system read/write/delete (os), network access (urllib), and shell command execution (subprocess).
- Sanitization: No sanitization or validation of the Markdown content or generated titles is performed before processing.
Recommendations
- HIGH: Downloads and executes remote code from: https://astral.sh/uv/install.sh - DO NOT USE without thorough review
Audit Metadata