---
type: "post"
id: "persistent-memory-for-claude-code-with-weaviate-engram-75aa8e"
slug: "persistent-memory-for-claude-code-with-weaviate-engram-75aa8e"
title: "Persistent memory for Claude Code with Weaviate Engram"
description: "**Thesis:** Engram makes project history a cross-session context layer for Claude Code, rather than limiting retained notes to an individual repository."
retrieval_nugget: "**Thesis:** Engram makes project history a cross-session context layer for Claude Code, rather than limiting retained notes to an individual repository."
published_at: "2026-10-08"
updated_at: "2026-10-08"
record_date: "2026-10-08"
date_kind: "published_at"
topics: ["claude","claude-code"]
source_urls: ["https://weaviate.io/blog/engram-for-claude-code"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:persistent-memory-for-claude-code-with-weaviate-engram-75aa8e"
status: "published"
visuals: []
editorial_provenance: {"schema_version":"newruntime-editorial-copy-v1","content_status":"source_grounded_final","final_copy_sha256":"sha256:be0fe746161753440afa7a86f558c8e2a9fd87ec373444278fc2ef14cbc38ef0","reviewed_at":"2026-10-08T21:00:35.359Z","source_evidence_count":1,"verified_claim_count":3,"site_analysis_schema_version":"newruntime-site-analysis-v1","site_object_kind":"field_note","observed_fact_count":3,"implication_count":2,"watch_condition_count":1,"related_record_count":0}
analysis: {"schema_version":"newruntime-site-analysis-v1","object_kind":"field_note","thesis":"Engram makes project history a cross-session context layer for Claude Code, rather than limiting retained notes to an individual repository.","observed_facts":[{"text":"Before Claude answers, the plugin searches Engram for memories related to the prompt and passes relevant results as context.","source_urls":["https://weaviate.io/blog/engram-for-claude-code"]},{"text":"After the answer, it saves the exchange asynchronously and turns conversation into distilled memory notes.","source_urls":["https://weaviate.io/blog/engram-for-claude-code"]},{"text":"Weaviate says team-shared memories are possible depending on how topics are configured, and lists a free allowance of up to 1,000 runs per month.","source_urls":["https://weaviate.io/blog/engram-for-claude-code"]}],"mechanism":"The plugin hooks into both the pre-answer and post-answer stages: retrieval supplies prior context, while asynchronous extraction and storage make new context available later. This design puts memory in the workflow as a side effect rather than requiring a user to issue a dedicated “remember this” command.","why_now":"Claude Code work commonly crosses sessions and repositories; the product announcement targets the gap between local project notes and decisions that need to travel with a developer’s work.","implications":["If recall is reliable, developers may spend less time re-explaining earlier decisions or rediscovering cross-repository interfaces.","Shared topics could also carry knowledge between team members, making scope and access controls consequential."],"evidence_boundary":"These are Weaviate’s product claims. The announcement supplies no independent evaluation of recall accuracy, privacy, or the quality of generated memories; the stated free tier and failure behavior are also company descriptions.","watch_conditions":["Watch for independent tests of whether recalled memories are accurate and appropriately scoped across repositories and team topics."],"related_records":[],"new_branch_reason":"No prior New Runtime record was supplied; this release starts a distinct analytical record about persistent cross-repository memory in Claude Code."}
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/persistent-memory-for-claude-code-with-weaviate-engram-75aa8e/","markdown":"https://newruntime.com/posts/persistent-memory-for-claude-code-with-weaviate-engram-75aa8e.md","json":"https://newruntime.com/posts/persistent-memory-for-claude-code-with-weaviate-engram-75aa8e.json"}
---

# Persistent memory for Claude Code with Weaviate Engram

## Retrieval answer

**Thesis:** Engram makes project history a cross-session context layer for Claude Code, rather than limiting retained notes to an individual repository.

**Thesis:** Engram makes project history a cross-session context layer for Claude Code, rather than limiting retained notes to an individual repository. The announcement describes a plugin for Weaviate’s Engram memory server that can retrieve and retain information across codebases.

**Observed facts:** Before Claude answers, the plugin searches Engram for memories related to the prompt and passes relevant results as context. After the answer, it saves the exchange asynchronously and turns conversation into distilled memory notes. Weaviate says team-shared memories are possible depending on how topics are configured, and lists a free allowance of up to 1,000 runs per month.

**Mechanism:** The plugin hooks into both the pre-answer and post-answer stages: retrieval supplies prior context, while asynchronous extraction and storage make new context available later. This design puts memory in the workflow as a side effect rather than requiring a user to issue a dedicated “remember this” command.

**Why now:** Claude Code work commonly crosses sessions and repositories; the product announcement targets the gap between local project notes and decisions that need to travel with a developer’s work.

**Implications:** If recall is reliable, developers may spend less time re-explaining earlier decisions or rediscovering cross-repository interfaces. Shared topics could also carry knowledge between team members, making scope and access controls consequential.

**Evidence boundary:** These are Weaviate’s product claims. The announcement supplies no independent evaluation of recall accuracy, privacy, or the quality of generated memories; the stated free tier and failure behavior are also company descriptions.

**Watch condition:** Watch for independent tests of whether recalled memories are accurate and appropriately scoped across repositories and team topics.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Claude](https://newruntime.com/topics/claude.md): Explore the claude topic hub.
- topic hub: [Claude Code](https://newruntime.com/topics/claude-code.md): Explore the claude-code topic hub.
- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)
