alpaca-broker-account-onboarding
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill documents an API integration that ingests user-provided information to create financial accounts, creating a potential vector for malicious data to influence agent behavior.
- Ingestion points: The
POST /v1/accountsrequest body described inSKILL.md(§2) which takes contact, identity, and agreement data. - Boundary markers: No specific delimiters or safety instructions are defined to separate untrusted user input from the API request structure.
- Capability inventory: Network writing operations (
POST,PATCH) to Alpaca API endpoints for account creation and document uploads. - Sanitization: Section 7 of
SKILL.mdrecommends normalizing inputs to canonical formats like ISO 3166-1 alpha-3, which provides basic validation but does not fully mitigate prompt injection. - [EXTERNAL_DOWNLOADS]: The skill references several external resources for documentation and API schemas.
- Evidence: URLs pointing to
docs.alpaca.markets/docs/accounts,docs.alpaca.markets/reference/createaccount, anddocs.alpaca.markets/llms.txtinSKILL.mdandreference.md. - Context: These references target the official documentation and metadata files of Alpaca Markets, the service provider for which this skill is designed.
Audit Metadata