skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/llm-data-automation/Gen Agent Trust Hub
llm-data-automation
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documentation includes instructions to install Ollama using a shell script fetched from ollama.com via curl. This is a standard setup procedure for the recommended local AI tool.
- [COMMAND_EXECUTION]: The skill utilizes shell commands for environment setup and model management, including pip install for dependencies and ollama CLI commands for pulling and running models.
- [INDIRECT_PROMPT_INJECTION]: The skill's workflow involves processing external data sources such as PDF, Excel, and CSV files provided by users. This ingestion of untrusted external content represents a potential surface for indirect prompt injection, although the instructions explicitly advise users to validate inputs and check generated code before execution.
- [SAFE]: No malicious patterns, obfuscation, or unauthorized data exfiltration behaviors were detected. The requested filesystem permissions and the documented network references to Ollama and LM Studio align with the stated purpose of providing a data automation utility.
Audit Metadata