obsidian-vault
Pass
Audited by Gen Agent Trust Hub on Mar 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides shell snippets using
findandgrepthat incorporate user-provided keywords. If the agent interpolates these keywords directly into the shell without escaping, it could lead to command injection (e.g., using a keyword containing;or&&). - [PROMPT_INJECTION]: The skill ingests untrusted data from external Markdown files into the agent's context, creating an indirect prompt injection surface.
- Ingestion points: Local files searched/read from
/mnt/d/Obsidian Vault/AI Research/. - Boundary markers: None specified to delimit retrieved note content from instructions.
- Capability inventory: Shell command execution (
bash) for searching and managing files. - Sanitization: No sanitization or validation of note content is described before it is processed by the agent.
Audit Metadata