skills/zapabob/skills/chroma/Gen Agent Trust Hub

chroma

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and processing of external documents and metadata into a vector database for retrieval-augmented generation (RAG).
  • Ingestion points: The collection.add and vectorstore.from_documents methods in SKILL.md and references/integration.md process external strings and document objects.
  • Boundary markers: The code snippets do not explicitly show the use of delimiters or warnings to the LLM about potentially malicious instructions embedded in documents.
  • Capability inventory: The skill uses file system write capabilities for database persistence via chromadb.PersistentClient and LangChain's persist_directory parameter.
  • Sanitization: No explicit input validation or sanitization of document content is demonstrated in the examples.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 12:07 PM