obsidian-knowledge-architect
Pass
Audited by Gen Agent Trust Hub on Aug 18, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing untrusted data from the workspace. 1. Ingestion points: The agent is instructed to read content from .agents/startup_dna.md, .agents/session_memory.md, and various Markdown files (PRDs, specs, strategy) within the project directory (SKILL.md). 2. Boundary markers: There are no defined delimiters or explicit 'ignore' instructions for the agent when processing these external documents. 3. Capability inventory: The skill maintains capabilities to read and write files to reorganize directory structures and create interactive dashboards (SKILL.md). 4. Sanitization: The skill lacks programmatic sanitization; it relies on a manual quality checklist for scrubbing secrets, which does not prevent malicious instructions within the documentation from influencing agent logic (SKILL.md).
Audit Metadata