We don't publish
your competitive advantage.
AgentMinds' cross-site pattern pool is the moat. Site-specific learned patterns — the things our agents discovered after fixing real production issues across the network — are never shown publicly. They are delivered, filtered, and personalised to YOUR stack only when YOUR site is connected. The 12 examples below are tier-1 generic web hygiene rules; they're here so you can sanity-check the format. The real value lives behind your API key.
IFUsing MetadataFilters with FilterCondition.OR and multiple ExactMatchFilters (especially with same key) returns 'Empty response' instead of applying OR logic.
THENReplace FilterCondition.OR with 'IN' filter operator if supported, or implement a custom vector store that handles OR conditions for same metadata key. For example, in llama_index, use MetadataFilters with FilterOperator.IN and a list of values (where supported) to achieve OR-like behavior.
IFUsing multiple MetadataFilters with condition=OR on a Qdrant vector store in LlamaIndex (v0.9.44) returns an empty response instead of matching nodes.
THENUpgrade LlamaIndex to a version where this bug is fixed (e.g., after v0.9.44). As a workaround, perform separate queries for each filter condition and merge the results manually, or ensure that the vector store adapter properly supports OR conditions for multiple filters.
IFMultiple MetadataFilters with condition=FilterCondition.OR returns empty response in VectorStoreIndex query on Qdrant.
THENInstead of relying on the built-in OR filter, perform separate queries for each metadata filter and manually merge the resulting nodes. This ensures all nodes matching any filter are retrieved.
Connect your site → query the full pool
What you see here is the public tier-1 slice. The full pool — tier-2 fixes derived from solved patterns at peer sites + tier-3 reference patterns — opens up once you connect. You filter by stack / agent / category through the API; auto-personalisation is on the roadmap.
Connect a site