new·The score now tells you which way it movedA brain's exam only ever grows: its own material writes questions, and so does every question a real caller asked and did not get answered. The score is a percentage over that growing set, so a brain that learned more could post a smaller number — and this week three did. One of them answered two MORE questions than the week before and showed eighteen points less. Printed as a single percentage, that reads as decline to a reader and as punishment to anyone who contributes material.all news →
mozg.beta
Sign in

LangChain · LangGraph · all subjects

memory & store

8 notes, read out of this brain and free to use. Each one was extracted from a source and is re-checked against its exam.

Store namespace matching behavior

Store search and list_namespaces scope namespace matching to segment boundaries and treat namespace labels as literal characters. The underscore (_) and percent (%) characters are treated as literal characters, not as wildcard characters.

gRPC store backend with TTL support

v0.12 adds gRPC-backed store backend supporting get and put operations with TTL, as well as delete, search, and list namespaces operations.

Custom encryption for Store API value field

Custom encryption is supported for the Store API `value` field, allowing users to choose which keys to encrypt for enhanced security.

get_store() for store access in custom routes

`get_store()` function enables Store access from user-defined Starlette endpoints, allowing store operations within custom routes.

Dataset schema prebuilt types in LangSmith

LangSmith provides prebuilt JSON schema types for dataset inputs and outputs to ensure data consistency and correct format for downstream processing like evals. These schemas are hosted publicly by the LangSmith API and can be referenced using JSON Schema references.

Message prebuilt schema type

The Message type has JSON Schema Reference Link https://api.smith.langchain.com/public/schemas/v1/message.json and represents messages sent to a chat model, following the OpenAI standard format.

Tool prebuilt schema type

The Tool type has JSON Schema Reference Link https://api.smith.langchain.com/public/schemas/v1/tooldef.json and represents tool definitions available to chat models for function calling, defined in OpenAI's JSON Schema inspired function format.

Dataset schema transformations

LangSmith lets you define a series of transformations that collect prebuilt types from your traces and add them to your dataset. More info on available transformations is available in the dataset transformations reference.

Give your agent this brain