obsidian-clipping-repair
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's workflow involves fetching content from external URLs (Step 3: Recover the true source; Step 4: Extract the article) and incorporating it into local Markdown notes, creating a vector for indirect prompt injection. \n
- Ingestion points: External web pages retrieved via agent browsing and existing vault Markdown files. \n
- Boundary markers: The instructions do not specify the use of delimiters or warnings when processing external content. \n
- Capability inventory: The agent has file read/write access to the Obsidian vault and web browsing capabilities. \n
- Sanitization: No explicit sanitization of external content is described. \n- [COMMAND_EXECUTION]: The skill utilizes a bundled Python script,
scripts/find_candidates.py, for vault triage. While the script is local and limited to standard libraries, it represents an execution point on the local filesystem.
Audit Metadata