albert-api
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and processing of untrusted external content, which is a known attack surface for indirect prompt injection.
- Ingestion points: The
SKILL.mdfile defines workflows for processing user messages viaPOST /v1/chat/completionsand indexing external documents (PDFs, images) viaPOST /v1/documentsfor RAG operations. - Boundary markers: The skill documentation does not provide instructions for the agent to use delimiters or specific safety framing when interpolating untrusted data into prompt contexts.
- Capability inventory: The skill enables the agent to perform network operations (HTTP requests to the Albert API) and manage sensitive data (API keys, user usage statistics) which could be influenced by malicious instructions embedded in processed data.
- Sanitization: There is no evidence of sanitization, filtering, or validation of the external content before it is processed by the agent or sent to the API endpoints.
Audit Metadata