Introducing the Semantic Cache Layer for Superagent 💯

In order to save tokens and inference speed we have implemented a semantic cache. This cache is by default set to revalidate after 24h but you can override it by setting cache_ttl to 0 when invoking the agent through the SDKs our API.

What's Changed

• Add warning message when no API token exists by @homanp in #317
• Semantic query cache by @simjak in #315

Full Changelog: v0.0.58...v0.0.59