obsidian-cli
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill operates on user-supplied data such as note content, search queries, and URI parameters, which creates a surface for indirect prompt injection where malicious instructions could be embedded in the processed data.
- Ingestion points: Untrusted data enters the agent context through CLI arguments in
scripts/run-command.shand URI strings inscripts/open-uri.sh. - Boundary markers: The scripts do not utilize specific delimiters or defensive prompts to distinguish user-provided content from automation instructions.
- Capability inventory: The skill has the capability to read and write vault files and execute application-level logic via the
obsidianbinary'sevalcommand, as well as trigger system-level URI actions. - Sanitization: No sanitization or input validation is performed on the strings before they are passed to shell execution or the
obsidianbinary. - [COMMAND_EXECUTION]: The skill provides helper scripts (
scripts/run-command.sh,scripts/open-uri.sh) that wrap shell execution of theobsidianbinary and platform-specific URI openers likeopen,xdg-open, andcmd.exe. These tools are required for the skill's primary purpose of Obsidian automation.
Audit Metadata