spot-drift
Pass
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local commands using the
obsidianCLI via Bash to retrieve file lists, read note content, and query metadata. These operations are limited to the user's local Obsidian vault folders (e.g., `2 - Areas/Goals/
,1 - Projects/`) and are consistent with the skill's stated purpose of drift analysis.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from user-generated Obsidian notes.
- Ingestion points: Vault data retrieved via
obsidian readandobsidian search:contextinSKILL.md. - Boundary markers: None identified; the skill does not explicitly use delimiters when processing note content.
- Capability inventory: The agent has the ability to execute
obsidianCLI commands and perform directory listings. - Sanitization: No explicit sanitization of note content is mentioned before the agent synthesizes the drift report. While this creates a surface for indirect injection, the risk is limited by the local nature of the data and the skill's restricted scope.
Audit Metadata