doc-learner
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's core functionality involves analyzing external, potentially untrusted documents (such as prompts, technical guides, and API documentation) to extract structural patterns and writing styles. This creates a surface for indirect prompt injection where malicious instructions hidden within the analyzed text could influence the agent's behavior.
- Ingestion points: Processes user-provided documents as described in the "Document Receiving and Classification" and "Deep Document Analysis" stages in
SKILL.md. - Boundary markers: The skill instructions do not explicitly define boundary markers or "ignore instructions" directives when processing external content.
- Capability inventory: The skill is primarily a text-to-text transformation tool; the YAML frontmatter does not request access to privileged tools (e.g., shell access, network, or file system writes).
- Sanitization: No specific sanitization or filtering of input documents is mentioned in the logic.
Audit Metadata