zotero-obsidian-bridge
Pass
Audited by Gen Agent Trust Hub on May 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script (
scripts/verify_paper_notes.py) to perform automated checks on the Markdown notes directory, validating frontmatter fields and header structures. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core workflow of ingesting external, untrusted data.
- Ingestion points: External Zotero metadata, annotations, and PDF full-text content are read into the agent's context (referenced in
references/WORKFLOW.md). - Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions to prevent the agent from obeying commands embedded within the literature being processed.
- Capability inventory: The skill has the capability to write to the local filesystem (
Papers/*.md,Knowledge/*.md) and execute Python scripts (scripts/verify_paper_notes.py). - Sanitization: There is no evidence of content sanitization or filtering logic to neutralize potential instructions hidden within the source papers or web items.
Audit Metadata