Geek-skills-mineru-pdf-parser
Pass
Audited by Gen Agent Trust Hub on May 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for executing the mineru CLI and mineru-api commands to process documents and start a local API server. The API server documentation suggests binding to 0.0.0.0, which exposes the service to the local network beyond the host machine.
- [EXTERNAL_DOWNLOADS]: The skill utilizes the mineru-models-download utility to fetch required AI models from established repositories including Hugging Face and Modelscope.
- [PROMPT_INJECTION]: The skill ingests and parses external documents from local paths or remote URLs, creating a surface for indirect prompt injection where instructions embedded within the processed documents could influence the agent's behavior.
- Ingestion points: The file_sources parameter in the parse_documents tool (found in SKILL.md and references/mineru_v3_reference.md).
- Boundary markers: None identified; the instructions do not specify the use of delimiters or warnings to treat document content as untrusted.
- Capability inventory: The skill utilizes subprocess execution for the mineru CLI and manages file system operations for document input and output.
- Sanitization: None identified; content extracted from documents is converted to Markdown without explicit filtering or sanitization steps mentioned.
Audit Metadata