audit-para
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
obsidianCLI via Bash to interact with the local Obsidian vault. It executes commands to list files, read properties, read content, append content, and move files. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the analysis of vault content.
- Ingestion points: Note content and frontmatter are ingested via
obsidian readandobsidian property:readinSKILL.mdduring the Phase 1 audit pass. - Boundary markers: No explicit delimiters or boundary markers (e.g., XML tags or clear 'ignore embedded instructions' warnings) are used when reading and processing note content.
- Capability inventory: The skill possesses significant capabilities, including file reading, file moving (
obsidian move), content appending (obsidian append), and file creation/overwriting (obsidian create ... overwrite). - Sanitization: There is no evidence of sanitization or filtering of note content before it is processed by the agent to generate the audit report.
Audit Metadata