obsidian-source-ingestion
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions facilitate the ingestion of untrusted external content into a project-local knowledge base, creating a potential surface for indirect prompt injection where malicious instructions in the source material could influence the agent during processing.
- Ingestion points: External materials are ingested into target Source/ folders as defined in SKILL.md.
- Boundary markers: Absent; the guidelines do not instruct the agent to use delimiters or ignore embedded instructions in the source content.
- Capability inventory: The skill instructions imply file-system write operations to update index/registry files and create notes in SKILL.md.
- Sanitization: Absent; no sanitization or filtering of external input is required by the instructions.
Audit Metadata