upstash-redis

Pass

Audited by Gen Agent Trust Hub on Sep 1, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill documents the official Upstash Redis SDK, which is a well-known service for serverless and edge runtimes. All referenced URLs target official Upstash documentation and repositories.
  • [SAFE]: The instructions explicitly warn against hardcoding the UPSTASH_REDIS_REST_TOKEN and correctly advocate for the use of environment variables or secret managers for credential storage.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides an interface to ingest data from an external Redis database, creating a potential data ingestion surface.
  • Ingestion points: Data is retrieved via redis.get() calls as shown in SKILL.md.
  • Boundary markers: The SDK uses automatic serialization and deserialization of native JavaScript types.
  • Capability inventory: The skill is scoped to Redis database operations (GET, SET, etc.) and does not include subprocess execution, file system writes, or arbitrary shell access.
  • Sanitization: The client handles serialization automatically, reducing the risk of manual parsing errors, and the skill suggests namespacing keys to manage the keyspace securely.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 1, 2026, 07:26 PM
Security Audit — agent-trust-hub — upstash-redis