langchain-ollama
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill integrates local LLMs with tool-calling capabilities, creating a potential surface for indirect prompt injection.\n
- Ingestion points: The
invoke()andstream()methods forChatOllamaandOllamaLLMinSKILL.mdprocess text input and message lists that may originate from untrusted sources.\n - Boundary markers: The provided usage examples do not include boundary delimiters or system instructions designed to prevent the model from obeying instructions embedded within processed data.\n
- Capability inventory: The skill allows the agent to communicate with local or remote Ollama servers via HTTP and enables the binding of Python functions as executable tools.\n
- Sanitization: The skill's implementation examples do not demonstrate input sanitization or validation to mitigate injection attacks.\n- [EXTERNAL_DOWNLOADS]: Fetches the
langchain-ollamapackage from the official Python Package Index (PyPI) and references model downloads from Ollama's official library.\n- [COMMAND_EXECUTION]: Provides instructions to executeollamaCLI commands for model management and server operation.
Audit Metadata