Skills
Navigate
skills/sickn33/agentic-awesome-skills/vector-database-ops/Gen Agent Trust Hub

vector-database-ops

Pass

Audited by Gen Agent Trust Hub on Oct 28, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill provides templates for constructing database queries based on external data, creating a potential surface for indirect prompt injection.\n
  • Ingestion points: Data enters the agent's context through query_embedding vectors and metadata filter values (such as tenant_id and doc_type) used in the Python and SQL snippets in SKILL.md.\n
  • Boundary markers: The provided code templates do not include specific delimiters or instructions to the agent to disregard instructions potentially embedded within the data.\n
  • Capability inventory: The skill describes operations involving docker for service deployment, the qdrant_client Python library for database operations, and SQL execution via the pgvector extension in SKILL.md.\n
  • Sanitization: The example snippets do not demonstrate input sanitization or schema validation for the processed metadata values.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 28, 2026, 02:19 AM