zotero-obsidian-bridge
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources (Zotero metadata, annotations, and full-text PDFs) to populate an Obsidian knowledge base.\n
- Ingestion points: Data is pulled via Zotero tools as described in
references/WORKFLOW.md.\n - Boundary markers: The skill uses strict markdown schemas (
references/PAPER-NOTE-SCHEMA.md) and an 'Evidence Record' format to structure external data. It does not use explicit safety delimiters.\n - Capability inventory: The skill performs file writes to create/update Markdown and Obsidian Canvas files, and it executes a local Python script for verification.\n
- Sanitization: The 'Claim Promotion Gate' serves as a logic-based filter to prevent weak or unverified sources from influencing synthesized knowledge notes.\n- [COMMAND_EXECUTION]: The workflow includes a step to execute a local Python script for note verification.\n
- Evidence:
references/WORKFLOW.mdsuggests runningpython3 "${CLAUDE_PLUGIN_ROOT}/skills/zotero-obsidian-bridge/scripts/verify_paper_notes.py".\n - The script
scripts/verify_paper_notes.pyis part of the skill package and performs regex-based checks on the vault's content.
Audit Metadata