obsidian-rlm-distiller
Pass
Audited by Gen Agent Trust Hub on Jun 12, 2026
Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes external wiki source files, creating an attack surface for indirect prompt injection where malicious content in the sources could attempt to influence LLM behavior. * Ingestion points: Wiki source files are read and processed by the distill_wiki.py script. * Boundary markers: The documentation does not specify the use of delimiters or isolation prompts to protect against instructions embedded in source data. * Capability inventory: The skill has access to Read/Write tools and Bash for executing CLI commands. * Sanitization: No input validation or content sanitization is documented.\n- [DATA_EXFILTRATION]: Local wiki data is transmitted to external cloud LLM services including GitHub Copilot, Anthropic Claude, and Google Gemini. These are well-known services used as intended for the skill's primary summary-generation purpose.\n- [COMMAND_EXECUTION]: The skill executes local Python orchestration scripts and third-party LLM CLI tools via the Bash tool to perform its automated distillation tasks.
Audit Metadata