session-notes
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
obsidianCLI to perform vault operations such as creating notes (obsidian create), appending content (obsidian append), and searching (obsidian search). It includes a fallback to theWritetool if the CLI is unavailable and requires user confirmation before any write operation. - [EXTERNAL_DOWNLOADS]: The
captureworkflow allows fetching content from user-provided URLs using tools likeWebFetch. This is used to gather information to be saved into the Obsidian vault. - [PROMPT_INJECTION]: The skill identifies and mitigates indirect prompt injection risks when processing external content. (1) Ingestion points: Untrusted data is ingested via URLs and file attachments in
references/capture.md. (2) Boundary markers: The skill structures content into defined markdown sections and frontmatter. (3) Capability inventory: The skill possesses the ability to write to the local file system using both the Obsidian CLI and theWritetool. (4) Sanitization: The skill contains explicit instructions inSKILL.mdandreferences/capture.mdto treat fetched data solely as reference material and to discard any embedded directives.
Audit Metadata