api-vector-db-chroma
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill recommends the installation of official and well-known NPM packages for the Chroma ecosystem, including
chromadband various@chroma-coreutility packages. - [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and retrieval of document data, creating a surface for indirect prompt injection attacks where malicious instructions could be embedded in stored content.
- Ingestion points: Document content is ingested via
collection.add,collection.upsert, andcollection.updatemethods acrossSKILL.mdandexamples/core.md. - Boundary markers: The provided patterns do not include explicit delimiters or instructions to ignore embedded commands within document text.
- Capability inventory: The skill performs network operations to connect to Chroma servers and potentially third-party embedding providers, and it modifies the database state.
- Sanitization: While metadata is validated for flatness, there is no evidence of sanitization or escaping for the
documentscontent to prevent prompt injection.
Audit Metadata