obsidian-clipping-ai-summary
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external content, such as Bilibili video transcripts and meeting notes, to generate summaries and sync them to Feishu. It lacks explicit boundary markers or sanitization protocols to prevent embedded instructions in these sources from being executed by the agent. 1. Ingestion points: Obsidian note sections (e.g., ## 笔记, ## 字幕). 2. Boundary markers: Absent. 3. Capability inventory: File system write/rename access and Feishu API synchronization. 4. Sanitization: Absent (residue removal is for formatting purposes only).
- [COMMAND_EXECUTION]: The skill instructs the agent to utilize CLI utilities like 'sed' and 'rg' (ripgrep) to verify the results of its file modifications. Executing commands on files that contain or are derived from untrusted external data presents a surface for command-line manipulation.
Audit Metadata