didit-proof-of-address
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a standard API integration for document verification. All referenced URLs (verification.didit.me, apx.didit.me, docs.didit.me) belong to the authorized vendor infrastructure of didit-protocol.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted documents provided by the user, which presents an attack surface for indirect prompt injection if the document contains malicious instructions. However, the risk is minimal in this context because the tool returns structured data rather than raw text, and the skill lacks capabilities that could be exploited by such an injection.
- Ingestion points: The skill accepts user-uploaded files (images/PDFs) via the
documentparameter inscripts/verify_address.py. - Boundary markers: The API returns a structured JSON response (e.g.,
poa.status,poa.poa_parsed_address), which helps demarcate data from instructions. - Capability inventory: The skill is restricted to performing HTTP POST requests to a specific verification endpoint; it does not execute arbitrary shell commands or local modifications based on document contents.
- Sanitization: The Didit API manages the extraction and geocoding, serving as a buffer between the raw file content and the agent context.
Audit Metadata