# gralkor_ex v2.1.4 - Table of Contents

Embedded Gralkor memory for Elixir/OTP — runs Graphiti + FalkorDB in-process via PythonX. Embed in a Jido (or any Elixir) supervision tree to give your agent long-term, temporally-aware knowledge-graph memory.

## Pages

- [gralkor_ex](readme.md)

## Modules

- [Gralkor.CaptureBuffer](Gralkor.CaptureBuffer.md): In-flight conversation buffer keyed by `session_id`.
- [Gralkor.Client](Gralkor.Client.md): Port for talking to a Gralkor backend from Elixir.
- [Gralkor.Client.InMemory](Gralkor.Client.InMemory.md): In-memory twin of `Gralkor.Client` for tests.
- [Gralkor.Client.Native](Gralkor.Client.Native.md): Production `Gralkor.Client` implementation. In-process — no HTTP — talks
to graphiti via `Gralkor.GraphitiPool` (Pythonx-backed) and to the LLM via
`req_llm` (Elixir-side, used by `Gralkor.Distill` and `Gralkor.Interpret`).
- [Gralkor.Config](Gralkor.Config.md): Configuration for the embedded Gralkor runtime.
- [Gralkor.Distill](Gralkor.Distill.md): Render a list of conversation turns into an episode body suitable for
ingesting into the knowledge graph.
- [Gralkor.Format](Gralkor.Format.md): Format graphiti edge data into the canonical fact strings the LLM sees.
- [Gralkor.GraphitiPool](Gralkor.GraphitiPool.md): Per-group Graphiti instance cache, plus the gateway for graphiti operations.
- [Gralkor.Interpret](Gralkor.Interpret.md): Filter retrieved graph facts down to those relevant to the conversation,
using the configured LLM.
- [Gralkor.Message](Gralkor.Message.md): Canonical message shape that Gralkor expects at its port boundary.
- [Gralkor.Python](Gralkor.Python.md): PythonX runtime owner for the embedded Gralkor stack.
- [Gralkor.Recall](Gralkor.Recall.md): Orchestrate one recall call: search the graph, interpret the hits against
the buffered conversation, wrap the result in a `<gralkor-memory>` block.

