server-connection-scala

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill utilizes the official and versioned Couchbase Scala SDK dependency (com.couchbase.client %% scala-client % 1.7.0).
  • [SAFE]: The primary connection examples correctly demonstrate using environment variables (sys.env) to retrieve sensitive credentials, avoiding hardcoded secrets in production code.
  • [SAFE]: A placeholder password ('Password!123') is used in a Capella connection example for illustrative purposes. This is identified as a documentation placeholder rather than a hardcoded credential leak.
  • [PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection as it facilitates the ingestion of data from an external database into the agent's context.
  • Ingestion points: Data is retrieved via collection.get and collection.lookupIn in SKILL.md and examples/examples.md.
  • Boundary markers: No specific boundary markers or 'ignore' instructions are suggested for the data retrieved from the cluster.
  • Capability inventory: The skill allows for high-privilege operations including upsert, remove, and mutateIn.
  • Sanitization: No explicit sanitization or schema validation logic is demonstrated for the content retrieved from the database.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 09:29 AM
Security Audit — agent-trust-hub — server-connection-scala