AutoGineer
Generated banner for LongMemory
Generated logo for LongMemory

LongMemory

Memory server

Created by CaviraOSS

Published in CaviraOSS/LongMemory. Publisher page

4,573 GitHub stars

What it is

LongMemory Hydrograph TypeScript package: library, CLI, self-hosted API server, and benchmark harness.
— the description by CaviraOSS

AutoGineer overview, from the brain’s own files

LongMemory is a cognitive memory engine designed to provide AI agents with durable, governed, and temporal recall. Unlike standard RAG pipelines that rely on simple vector similarity, LongMemory utilizes a Hydrograph substrate to manage immutable nodes, executable edges, and temporal truth. It supports complex recall modes including strict, historical, associative, and world-grounded retrieval. The system is built for local-first deployment using SQLite, offering a self-hosted API server, CLI, and native integrations for various agent frameworks, IDEs, and automation tools. It enforces governance policies across projects, users, and teams, ensuring that memory remains grounded, explainable, and context-aware.

How it works

Memory model: LLM wiki

AutoGineer overview, from the brain’s own files

LongMemory operates as a memory server that stores data in a local SQLite database. It ingests events, documents, and session data into a Hydrograph structure. When an agent needs information, the system performs recall based on specific modes—such as strict, historical, or associative—to retrieve relevant, token-bounded context. The engine supports governance by scoping memory to specific projects, tenants, or users. It provides an API for agents to interact with this memory, and includes tools for maintenance, such as decay and reinforcement, to manage the lifecycle of stored information.

Folders
FolderRoleFiles
.githubagent-config13
docsdocs34
packagessource-code7
srcsource-code340
toolsscripts5
.agentsfolder1
.dofolder1
appsfolder19
benchmarksfolder32
dashboardfolder34
integrationsfolder47

Requirements

Apps: Docker

Runtimes: Node.js >=20

Databases: SQLite

API keys (by name): AWS_SECRET_ACCESS_KEY, AWS_SESSION_TOKEN, AZURE_DEVOPS_TOKEN, BITBUCKET_TOKEN, GEMINI_API_KEY, GITHUB_TOKEN, GITLAB_TOKEN, GOOGLE_ACCESS_TOKEN, LINEAR_API_KEY, LONGMEMORY_API_KEY, MICROSOFT_GRAPH_TOKEN, NOTION_API_KEY, NVIDIA_API_KEY, OM_API_KEY, OM_GEMINI_API_KEY, OM_OPENAI_API_KEY, OM_SIRAY_API_TOKEN, OPENAI_API_KEY, SIRAY_API_TOKEN, SLACK_BOT_TOKEN

Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Google Antigravity, Kiro, OpenCode, Amp, Roo Code, Cline, JetBrains Junie, Devin (ex-Windsurf)

Set it up

  1. This brain runs code automatically: 1 script. Review Makefile before opening the folder in Claude Code. No step below turns a hook on for you.

    • Makefile: corepack enable; pnpm install --frozen-lockfile
  2. Not scanned by AutoGineerThe author’s README also describes installing it with Docker Compose. Read its install section: AutoGineer did not scan that path. https://github.com/CaviraOSS/LongMemory/blob/aec98ced33a7586a2fcd198a04bc38b5590750c4/README.md#2-run-as-a-service
  3. Not scanned by AutoGineerThe author’s README also describes installing it from a clone of its repository. Read its install section: AutoGineer did not scan that path. https://github.com/CaviraOSS/LongMemory/blob/aec98ced33a7586a2fcd198a04bc38b5590750c4/README.md#2-run-as-a-service

  4. Created by CaviraOSS. Published in CaviraOSS/LongMemory. Licence: Apache License 2.0.

Usage

AutoGineer overview, from the brain’s own files

  • Install as a TypeScript library for in-memory or SQLite-backed agent memory.
  • Deploy as a self-hosted API server using Docker or Docker Compose.
  • Use the CLI to ingest data, manage project context, or perform maintenance tasks.
  • Connect to agent hosts via MCP (Model Context Protocol) for seamless integration.
  • Use the Python SDK to interact with the memory server from Python-based agents.
  • Visualize and manage memory through the provided Next.js dashboard or VS Code extension.
“What was the deployment region in January?”
The agent uses historical recall mode with a specified valid_time to retrieve the correct configuration from that period.
“Prepare the next release.”
The agent retrieves project context, including architecture notes, open tasks, and relevant code facts, to assist with release planning.

Documentation by CaviraOSS

<!-- __ __ ___ / / ____ ____ ____ _/ |/ /__ ____ ___ ____ _______ __ / / / __ \/ __ \/ __ `/ /|_/ / _ \/ __ `__ \/ __ \/ ___/ / / / / /___/ /_/ / / / / /_/ / / / / __/ / / / / / /_/ / / / /_/ / /_____/\____/_/ /_/\__, /_/ /_/\___/_/ /_/ /_/\____/_/ \__, / /____/ /____/ cavira oss (c) 2026 - nullure (c) 2026 ========================================================== file : README.md usage : introduces LongMemory, its architecture, integrations, and deployment options -->

LongMemory

Durable, temporal, governed memory for AI agents. Not just RAG. Not just a vector database. Local-first and self-hosted.

Image: npm Image: PyPI Image: VS Code Image: Container Image: License

Image: LongMemory dashboard

LongMemory is a cognitive memory engine for LLM applications and autonomous agents.

  • Durable local-first storage with SQLite

  • Immutable content, provenance, and temporal truth

  • Strict, historical, associative, grounded, and multilingual recall

  • Explainable evidence selection and token-bounded context

  • Governed project memory, Skills, Chat Memory, LLM-Wiki, and CodeGraph

  • One TypeScript engine across npm, CLI, HTTP, MCP, dashboard, and VS Code

  • Native integrations for agent hosts, automation tools, and Python frameworks

Your model stays stateless. Your application stops being amnesiac.

1. Use It in 10 Seconds

Install as a library
npm install longmemory
import { createMemory } from 'longmemory';

const memory = await createMemory();
await memory.ingest({
    user_id: 'alice',
    text: 'I prefer TypeScript for backend services',
});

const result = await memory.recall({
    text: 'What language does Alice prefer?',
    mode: 'strict',
});

console.log(result);
await memory.close();

No service or external database is required for in-memory use.

For optional evidence-grounded answers with your own model, use the exported answer_from_evidence adapter. It accepts authorized evidence, validates cited excerpts, and makes at most one model call; ordinary recall is unchanged. See docs/answering.md for policies, limits, and integration.

Persist with SQLite
const memory = await createMemory({
    store: 'sqlite',
    db_path: './longmemory.db',
    tenant_id: 'acme',
    user_id: 'alice',
});

Reopening the same database restores nodes, worlds, entities, edges, temporal history, grounding, and lifecycle state.

Install the CLI
npm install --global longmemory
longmemory init
longmemory recall "current project priorities" --mode associative
Call a self-hosted server from Python
pip install longmemory-sdk
from longmemory import LongMemory

memory = LongMemory(
    "http://127.0.0.1:7331",
    api_key="change-me",
    user_id="alice",
)

memory.ingest("I prefer TypeScript")
result = memory.recall("What language do I prefer?", mode="strict")

The Python package is a zero-dependency HTTP client. The Hydrograph engine remains in the self-hosted TypeScript service. See docs/python-sdk.md.

2. Run as a Service

From source
git clone https://github.com/CaviraOSS/LongMemory.git
cd LongMemory
corepack enable
pnpm install --frozen-lockfile
pnpm build
pnpm start

The API listens on http://127.0.0.1:7331 by default.

Docker
docker run --rm \
  -p 7331:7331 \
  -v longmemory-data:/data \
  -e LONGMEMORY_API_KEY=change-me \
  ghcr.io/caviraoss/longmemory:latest
Docker Compose
cp .env.example .env
docker compose up --build -d longmemory

Include the dashboard:

docker compose --profile ui up --build -d

  • API and MCP: http://127.0.0.1:7331

  • Dashboard: http://127.0.0.1:3000

  • Health: http://127.0.0.1:7331/health

3. Why LongMemory

Most systems called memory are retrieval pipelines:

  1. Split text into chunks.

  2. Embed the chunks.

  3. Return the nearest vectors.

That does not establish what was true at a particular time, whether a new fact superseded an old one, which source is authoritative, who may see it, or why a result belongs in context.

LongMemory models those concerns directly:

  • Temporal truth: recorded time and valid time are separate.

  • Immutable memory: content, vectors, hashes, and provenance are not rewritten by recall or decay.

  • Executable graph: typed relationships participate in recall and explanation.

  • Governance: project, tenant, user, team, role, agent, task, and framework scope are enforced.

  • Lifecycle: deterministic decay, explicit reinforcement, consolidation, compression, and reconsolidation.

  • Evidence: recall is bounded by relevance, contradictions, grounding, permissions, and token cost.

See Why.md for the design rationale.

4. Recall Modes

const strict = await memory.recall({
    text: 'What is the current deployment region?',
    mode: 'strict',
});

const historical = await memory.recall({
    text: 'What was the deployment region in January?',
    mode: 'historical',
    valid_time: Date.UTC(2026, 0, 15),
});

const associative = await memory.recall({
    text: 'Incidents related to the payment migration',
    mode: 'associative',
});

const grounded = await memory.recall({
    text: 'Which production endpoint is currently live?',
    mode: 'world_grounded',
});

Strict recall applies temporal, contradiction, contract, confidence, and grounding gates. Historical recall preserves superseded truth. Associative recall follows semantic, lexical, entity, activation, and graph signals. World-grounded recall requires current external evidence.

5. Features

  • Hydrograph memory substrate with immutable nodes, executable edges, worlds, entities, facets, and traces.

  • Temporal reasoning with point-in-time truth, event ordering, supersession, and stale-evidence controls.

  • Multilingual memory with script detection, code switching, transliteration, and cross-language embeddings.

  • Project memory for architecture, decisions, tasks, conventions, failures, handoffs, and code impact.

  • Governed assets for Chat Memory, Skills, LLM-Wiki, and CodeGraph with lifecycle and ACL policy.

  • Session porter for Claude Code, Codex, OpenCode, Gemini CLI, Copilot Chat, Cline, and raw harness logs.

  • Connectors for repositories, local files, Markdown, web content, feeds, cloud documents, and provider APIs.

  • Embeddings through OpenAI-compatible APIs, Gemini, AWS Bedrock, Ollama, Siray, and local HTTP models.

  • Operational surfaces through HTTP, MCP, dashboard, VS Code, n8n, and framework-native MCP clients.

  • Auditable benchmarks for LongMemEval, LoCoMo, BEAM, retrieval quality, temporal behavior, and latency.

6. MCP and Agent Integrations

Start local stdio MCP:

longmemory mcp --db .longmemory/project.db --project current

Expose authenticated Streamable HTTP MCP:

LONGMEMORY_API_KEY=change-me longmemory serve --mcp-http

LongMemory exposes 13 high-level governed tools plus readable resources and agent workflow prompts. Tool arguments cannot override server-bound runtime identity.

Installable integrations include:

  • Claude Code plugin

  • Codex and ChatGPT desktop plugin

  • Gemini CLI extension

  • Agent Plugins 1.0 bundle for OpenClaw and compatible hosts

  • n8n community node usable as an AI Agent tool

  • Cline, Continue, and LibreChat configuration packs

  • Dify and Flowise native MCP setup

  • CrewAI, AutoGen, LangGraph/LangChain, OpenAI Agents SDK, and PydanticAI examples

See integrations/README.md and docs/mcp.md.

7. Temporal and Project Memory

import { createProjectMemory } from 'longmemory';

const projects = await createProjectMemory({
    tenant_id: 'cavira',
    organization_id: 'CaviraOSS',
    project_id: 'longmemory',
    name: 'LongMemory',
    store: 'sqlite',
    db_path: './longmemory.db',
});

await projects.ingestProjectEvent('longmemory', {
    kind: 'decision',
    topic: 'persistence',
    text: 'Use SQLite for local-first persistence',
    source_type: 'architecture_note',
});

const context = await projects.getProjectContext('longmemory', 'prepare the next release');

Project context combines relevant architecture, current decisions, open tasks, failures, code facts, matched Skills, conflicts, and governed asset loadouts under one token budget.

8. CLI

longmemory init
longmemory tui
longmemory status --memories 20 --json
longmemory ingest "Remember the rollback procedure" --type procedure
longmemory recall "What is the rollback procedure?" --mode associative
longmemory memory list --limit 50
longmemory project context "prepare the next release"
longmemory maintenance decay --all
longmemory maintenance reinforce <memory-id>
longmemory skill match "run the release checklist" --agent reviewer
longmemory asset loadout "prepare the release" --agent reviewer --framework codex
longmemory code impact createMemory
longmemory detect
longmemory session discover --from claude-code
longmemory port --from claude-code --to longmemory --all
longmemory session wiki --from gemini-cli --all --name "Project knowledge"
longmemory serve --mcp-http

Finite commands emit stable JSON outside a TTY or when --json is supplied. The session porter reads supported coding-agent stores without modifying them. See docs/cli.md.

9. Dashboard and VS Code

The Next.js dashboard provides health, memory browsing, ingestion, search, project selection, activity, decay, settings, timelines, and memory-aware chat through a same-origin API proxy.

pnpm --dir dashboard build
pnpm --dir dashboard start

The VS Code extension provides an activity-bar browser, status bar, recall, project context, explanation, reinforcement, explicit decay, session import, and reviewed AI-change capture.

pnpm extension:package

The generated package is apps/vscode-extension/longmemory-vscode-0.2.0.vsix.

10. Architecture

graph TB
  INPUT[Events, documents, sessions] --> INGEST[Immutable ingest pipeline]
  INGEST --> GRAPH[(Hydrograph)]
  GRAPH --> STRICT[Strict and historical recall]
  GRAPH --> ASSOC[Associative recall]
  GRAPH --> GROUND[World-grounded recall]
  GRAPH --> PROJECT[Project memory and governed assets]
  GRAPH --> SQLITE[(SQLite)]
  STRICT --> CONTEXT[Explainable bounded context]
  ASSOC --> CONTEXT
  GROUND --> CONTEXT
  PROJECT --> MCP[MCP tools, resources, prompts]
  CONTEXT --> API[Library, CLI, HTTP]
  MCP --> AGENTS[Agents, IDEs, automation]
  API --> UI[Dashboard and VS Code]

Read ARCHITECTURE.md and docs/architecture.md for subsystem details.

11. Deployment Options

PlatformConfigurationWhat it deploys
DockerDockerfileAPI and Streamable HTTP MCP
Docker Composedocker-compose.ymlAPI/MCP plus optional dashboard
Herokuapp.jsonContainerized API/MCP
Railwayrailway.jsonContainerized API/MCP
Renderrender.yamlAPI/MCP with persistent disk
DigitalOcean.do/spec.yamlApp Platform API/MCP service
Vercelvercel.jsonDashboard; configure LONGMEMORY_API_URL
Windowsstart-longmemory.ps1Background local API/MCP process

For hosted API deployments, set LONGMEMORY_API_KEY, mount persistent storage at /data, and terminate TLS at the platform edge. Vercel hosts only the stateless dashboard and requires a separately deployed LongMemory API.

12. Benchmarks

pnpm bench
pnpm bench:ci
pnpm bench:full

The benchmark harness publishes explicit manifests, dataset completion, evidence metrics, answer judgments, temporal categories, latency percentiles, and N/A reasons. Official scorecards fail closed on incomplete datasets or semantic embedding fallback. See benchmarks/README.md.

13. Migration

Import supported SQLite, JSON, or JSONL memory:

longmemory migrate \
  --from ./legacy.db \
  --to ./longmemory.db \
  --report ./migration-report.json

Import coding-agent conversations as governed Chat Memory:

longmemory port --from codex --to longmemory --all

See MIGRATION.md and docs/migration.md.

Legacy package migration:

npm uninstall openmemory-js && npm install longmemory
pip uninstall openmemory-py && pip install longmemory-sdk

openmemory-js@2 and openmemory-py@2 are forwarding bridges for existing installations. New applications should use longmemory and longmemory-sdk directly. PyPI's unrelated longmemory name is owned by another project, so the official distribution is longmemory-sdk while the import remains longmemory.

14. Release and Operations

corepack enable
pnpm install --frozen-lockfile
pnpm release:check
pnpm pack
pnpm extension:package

release:check validates branding, types, integration manifests, the benchmark smoke gate, the root build, extension build, and dashboard production build.

Useful Make targets:

make install
make build
make check
make docker-up
make dashboard

15. Security

LongMemory is local-first, but network deployment still requires explicit controls:

  • Protect API and MCP routes with LONGMEMORY_API_KEY.

  • Restrict allowed origins and terminate TLS at the edge.

  • Keep connector and embedding credentials outside repository files.

  • Treat recalled content as untrusted evidence, not authorization.

  • Preserve server-bound user, project, agent, and framework identity.

Report vulnerabilities privately as described in SECURITY.md.

16. Contributing

Issues and pull requests are welcome. Read CONTRIBUTING.md, GOVERNANCE.md, and CODE_OF_CONDUCT.md before contributing.

17. License

LongMemory is licensed under the Apache License 2.0. The separately published n8n community node uses MIT as required by n8n's strict package validator.

Contributors

<!-- readme: contributors -start -->

<table> <tbody> <tr> <td align="center"> <a href="https://github.com/nullure"> <img src="https://avatars.githubusercontent.com/u/81895400?v=4" width="100;" alt="nullure"/> <br /> <sub><b>Morven</b></sub> </a> </td> <td align="center"> <a href="https://github.com/dontbanmeplz"> <img src="https://avatars.githubusercontent.com/u/59851616?v=4" width="100;" alt="dontbanmeplz"/> <br /> <sub><b>Chis</b></sub> </a> </td> <td align="center"> <a href="https://github.com/amihos"> <img src="https://avatars.githubusercontent.com/u/35190548?v=4" width="100;" alt="amihos"/> <br /> <sub><b>Hossein Amirkhalili</b></sub> </a> </td> <td align="center"> <a href="https://github.com/DKB0512"> <img src="https://avatars.githubusercontent.com/u/23116307?v=4" width="100;" alt="DKB0512"/> <br /> <sub><b>DKB</b></sub> </a> </td> <td align="center"> <a href="https://github.com/stevo1403"> <img src="https://avatars.githubusercontent.com/u/34807365?v=4" width="100;" alt="stevo1403"/> <br /> <sub><b>Stephen Adebayo</b></sub> </a> </td> <td align="center"> <a href="https://github.com/mameikagou"> <img src="https://avatars.githubusercontent.com/u/116348059?v=4" width="100;" alt="mameikagou"/> <br /> <sub><b>mrlonely</b></sub> </a> </td> </tr> <tr> <td align="center"> <a href="https://github.com/haosenwang1018"> <img src="https://avatars.githubusercontent.com/u/167664334?v=4" width="100;" alt="haosenwang1018"/> <br /> <sub><b>Sense_wang</b></sub> </a> </td> <td align="center"> <a href="https://github.com/recabasic"> <img src="https://avatars.githubusercontent.com/u/102372274?v=4" width="100;" alt="recabasic"/> <br /> <sub><b>Elvoro</b></sub> </a> </td> <td align="center"> <a href="https://github.com/zfaustk"> <img src="https://avatars.githubusercontent.com/u/4340287?v=4" width="100;" alt="zfaustk"/> <br /> <sub><b>Clio</b></sub> </a> </td> <td align="center"> <a href="https://github.com/suyua9"> <img src="https://avatars.githubusercontent.com/u/273297082?v=4" width="100;" alt="suyua9"/> <br /> <sub><b>suyua9</b></sub> </a> </td> <td align="center"> <a href="https://github.com/vincenzopalazzo"> <img src="https://avatars.githubusercontent.com/u/17150045?v=4" width="100;" alt="vincenzopalazzo"/> <br /> <sub><b>Vincenzo Palazzo</b></sub> </a> </td> <td align="center"> <a href="https://github.com/msris108"> <img src="https://avatars.githubusercontent.com/u/43115330?v=4" width="100;" alt="msris108"/> <br /> <sub><b>Sriram M</b></sub> </a> </td> </tr> <tr> <td align="center"> <a href="https://github.com/fparrav"> <img src="https://avatars.githubusercontent.com/u/9319430?v=4" width="100;" alt="fparrav"/> <br /> <sub><b>Felipe Parra</b></sub> </a> </td> <td align="center"> <a href="https://github.com/DoKoB0512"> <img src="https://avatars.githubusercontent.com/u/123281216?v=4" width="100;" alt="DoKoB0512"/> <br /> <sub><b>DoKoB0512</b></sub> </a> </td> <td align="center"> <a href="https://github.com/whiterabb17"> <img src="https://avatars.githubusercontent.com/u/90134688?v=4" width="100;" alt="whiterabb17"/> <br /> <sub><b>MistrHyde</b></sub> </a> </td> <td align="center"> <a href="https://github.com/octo-patch"> <img src="https://avatars.githubusercontent.com/u/266937838?v=4" width="100;" alt="octo-patch"/> <br /> <sub><b>Octopus</b></sub> </a> </td> <td align="center"> <a href="https://github.com/therexone"> <img src="https://avatars.githubusercontent.com/u/27387245?v=4" width="100;" alt="therexone"/> <br /> <sub><b>Ayush Singh</b></sub> </a> </td> <td align="center"> <a href="https://github.com/kishan0725"> <img src="https://avatars.githubusercontent.com/u/36665975?v=4" width="100;" alt="kishan0725"/> <br /> <sub><b>Kishan Lal</b></sub> </a> </td> </tr> <tr> <td align="center"> <a href="https://github.com/pc-quiknode"> <img src="https://avatars.githubusercontent.com/u/126496711?v=4" width="100;" alt="pc-quiknode"/> <br /> <sub><b>Peter Chung</b></sub> </a> </td> <td align="center"> <a href="https://github.com/muhammad-fiaz"> <img src="https://avatars.githubusercontent.com/u/75434191?v=4" width="100;" alt="muhammad-fiaz"/> <br /> <sub><b>Muhammad Fiaz</b></sub> </a> </td> <td align="center"> <a href="https://github.com/jasonkneen"> <img src="https://avatars.githubusercontent.com/u/502002?v=4" width="100;" alt="jasonkneen"/> <br /> <sub><b>Jason Kneen</b></sub> </a> </td> <td align="center"> <a href="https://github.com/Hchunjun"> <img src="https://avatars.githubusercontent.com/u/11238835?v=4" width="100;" alt="Hchunjun"/> <br /> <sub><b>shan</b></sub> </a> </td> <td align="center"> <a href="https://github.com/naabakkcrypto"> <img src="https://avatars.githubusercontent.com/u/258882739?v=4" width="100;" alt="naabakkcrypto"/> <br /> <sub><b>Naabakk</b></sub> </a> </td> <td align="center"> <a href="https://github.com/mikemikimike"> <img src="https://avatars.githubusercontent.com/u/186855910?v=4" width="100;" alt="mikemikimike"/> <br /> <sub><b>mikemikimike</b></sub> </a> </td> </tr> <tr> <td align="center"> <a href="https://github.com/mgajewskik"> <img src="https://avatars.githubusercontent.com/u/47600161?v=4" width="100;" alt="mgajewskik"/> <br /> <sub><b>Maciej Gajewski</b></sub> </a> </td> <td align="center"> <a href="https://github.com/ajitam"> <img src="https://avatars.githubusercontent.com/u/672661?v=4" width="100;" alt="ajitam"/> <br /> <sub><b>Matija Urh</b></sub> </a> </td> <td align="center"> <a href="https://github.com/machj8968-lab"> <img src="https://avatars.githubusercontent.com/u/274563644?v=4" width="100;" alt="machj8968-lab"/> <br /> <sub><b>machj8968-lab</b></sub> </a> </td> <td align="center"> <a href="https://github.com/buyua9"> <img src="https://avatars.githubusercontent.com/u/274236111?v=4" width="100;" alt="buyua9"/> <br /> <sub><b>buyua9</b></sub> </a> </td> <td align="center"> <a href="https://github.com/aziham"> <img src="https://avatars.githubusercontent.com/u/94932043?v=4" width="100;" alt="aziham"/> <br /> <sub><b>Hamza</b></sub> </a> </td> <td align="center"> <a href="https://github.com/GPT-AGI"> <img src="https://avatars.githubusercontent.com/u/181050363?v=4" width="100;" alt="GPT-AGI"/> <br /> <sub><b>Ting Zhang</b></sub> </a> </td> </tr> <tr> <td align="center"> <a href="https://github.com/RockinPaul"> <img src="https://avatars.githubusercontent.com/u/3741448?v=4" width="100;" alt="RockinPaul"/> <br /> <sub><b>Paul Zarudnev</b></sub> </a> </td> <td align="center"> <a href="https://github.com/oantoshchenko"> <img src="https://avatars.githubusercontent.com/u/1749531?v=4" width="100;" alt="oantoshchenko"/> <br /> <sub><b>Oleksandr Antoshchenko</b></sub> </a> </td> <td align="center"> <a href="https://github.com/lwsinclair"> <img src="https://avatars.githubusercontent.com/u/2829939?v=4" width="100;" alt="lwsinclair"/> <br /> <sub><b>Lawrence Sinclair</b></sub> </a> </td> <td align="center"> <a href="https://github.com/jungdaesuh"> <img src="https://avatars.githubusercontent.com/u/78460559?v=4" width="100;" alt="jungdaesuh"/> <br /> <sub><b>jungdaesuh</b></sub> </a> </td> <td align="center"> <a href="https://github.com/josephgoksu"> <img src="https://avatars.githubusercontent.com/u/6523823?v=4" width="100;" alt="josephgoksu"/> <br /> <sub><b>Joseph Goksu</b></sub> </a> </td> <td align="center"> <a href="https://github.com/EikoocS"> <img src="https://avatars.githubusercontent.com/u/80829508?v=4" width="100;" alt="EikoocS"/> <br /> <sub><b>EikoocS</b></sub> </a> </td> </tr> <tr> <td align="center"> <a href="https://github.com/Dhravya"> <img src="https://avatars.githubusercontent.com/u/63950637?v=4" width="100;" alt="Dhravya"/> <br /> <sub><b>Dhravya Shah</b></sub> </a> </td> <td align="center"> <a href="https://github.com/dflor003"> <img src="https://avatars.githubusercontent.com/u/103017?v=4" width="100;" alt="dflor003"/> <br /> <sub><b>Danil Flores</b></sub> </a> </td> <td align="center"> <a href="https://github.com/DAESA24"> <img src="https://avatars.githubusercontent.com/u/173488786?v=4" width="100;" alt="DAESA24"/> <br /> <sub><b>DAESA24</b></sub> </a> </td> <td align="center"> <a href="https://github.com/ammesonb"> <img src="https://avatars.githubusercontent.com/u/2522710?v=4" width="100;" alt="ammesonb"/> <br /> <sub><b>Brett Ammeson</b></sub> </a> </td> <td align="center"> <a href="https://github.com/auto-pr-bot"> <img src="https://avatars.githubusercontent.com/u/245575329?v=4" width="100;" alt="auto-pr-bot"/> <br /> <sub><b>auto-pr-bot</b></sub> </a> </td> <td align="center"> <a href="https://github.com/anush008"> <img src="https://avatars.githubusercontent.com/u/46051506?v=4" width="100;" alt="anush008"/> <br /> <sub><b>Anush</b></sub> </a> </td> </tr> <tr> <td align="center"> <a href="https://github.com/annelchavez-11594"> <img src="https://avatars.githubusercontent.com/u/271067077?v=4" width="100;" alt="annelchavez-11594"/> <br /> <sub><b>Annel Chavez</b></sub> </a> </td> <td align="center"> <a href="https://github.com/atao2004"> <img src="https://avatars.githubusercontent.com/u/148929819?v=4" width="100;" alt="atao2004"/> <br /> <sub><b>Anna Tao</b></sub> </a> </td> </tr> <tbody> </table> <!-- readme: contributors -end -->

CHANGELOG.md

<!-- __ __ ___ / / ____ ____ ____ _/ |/ /__ ____ ___ ____ _______ __ / / / __ \/ __ \/ __ `/ /|_/ / _ \/ __ `__ \/ __ \/ ___/ / / / / /___/ /_/ / / / / /_/ / / / / __/ / / / / / /_/ / / / /_/ / /_____/\____/_/ /_/\__, /_/ /_/\___/_/ /_/ /_/\____/_/ \__, / /____/ /____/ cavira oss (c) 2026 - nullure (c) 2026 ---------------------------------------------------------- file : CHANGELOG.md usage : records the complete OpenMemory-to-LongMemory project history -->

Changelog

All notable changes to LongMemory are documented here. The current project follows Semantic Versioning. The archived pre-rewrite releases are retained at the end of this file for historical context.

Unreleased

Python SDK and package registry migration

  • Added packages/longmemory-py, a zero-runtime-dependency Python HTTP client for the self-hosted LongMemory service.

  • Added synchronous LongMemory and asynchronous AsyncLongMemory clients.

  • Added typed convenience methods for health, ingest, recall, explain, worlds, entities, timeline, statistics, runtime information, and arbitrary API calls.

  • Added structured LongMemoryError and LongMemoryConnectionError failures that preserve API status, error code, and response metadata.

  • Kept the Hydrograph engine in TypeScript. The Python package is intentionally a transport client and does not duplicate persistence, retrieval, temporal, lifecycle, or governance logic.

  • Selected the PyPI distribution name longmemory-sdk because the unrelated longmemory project name is already registered by another organization. The Python import remains from longmemory import LongMemory.

  • Added a deprecated openmemory-py compatibility distribution that depends on longmemory-sdk and forwards the former Python import namespace.

  • Added a deprecated openmemory-js npm bridge that depends on and re-exports longmemory, while forwarding the former opm command to the new CLI.

  • Added migration documentation for npm and PyPI users moving from the former package names.

  • Expanded package publication automation for npm, PyPI, n8n, VS Code, and compatibility bridge releases.

Registry status

  • Reserved longmemory as the primary npm package name.

  • Preserved the existing openmemory-js npm channel as a migration bridge for users of the former JavaScript package.

  • Preserved the existing openmemory-py PyPI channel as a migration bridge for users of the former Python package.

  • Did not claim the unrelated longmemory PyPI project. New Python installs use longmemory-sdk.

Fixes

  • Notes with newline-separated preferences or facts now keep each claim in stored summaries without splitting wrapped lines into separate statements.

  • Embedded project-scoped writes. longmemory_ingest with a project_id, longmemory_remember_decision and connector imports go through applyImportPlan, which stored every node without a semantic vector, so those memories never received a vector score in recall (#216).

  • The HTTP server never read LONGMEMORY_TENANT_ID/LONGMEMORY_USER_ID, so every restart silently re-scoped the engine (and the /mcp endpoint) to the default tenant and user, making memories saved under a configured scope unreachable after a restart (#215).

  • MCP tool responses (longmemory_ingest, longmemory_recall, and others) embedded each node's full embedding vector, repeated across every place the node is nested in a response, inflating responses by thousands of tokens. Vectors are now stripped by default (LONGMEMORY_MCP_INCLUDE_VECTORS=true or a per-call include_vectors: true input restores them); ingest and recall responses also no longer repeat a node already returned in full elsewhere in the same response, and structuredContent — an unvalidated duplicate of content[].text, since no tool declares an outputSchema — is no longer emitted at all (#219).

Documentation

  • Added a guide for connecting Grok's Bring Your Own MCP connector to LongMemory's Streamable HTTP MCP endpoint, covering tunneling for local testing, a TLS-terminated deployment for persistent use, and authentication (integrations/grok-connectors.md) (#176).

1.0.0 - 2026-08-31

LongMemory 1.0 is a ground-up architecture rewrite rather than an incremental rename of the archived OpenMemory implementation. It consolidates the product around one immutable TypeScript Hydrograph engine shared by every supported surface.

Product rename and release identity

  • Renamed the product from OpenMemory to LongMemory.

  • Renamed the npm package and executable to longmemory.

  • Renamed public environment variables to the LONGMEMORY_* namespace.

  • Renamed workspace state from .openmemory/ to .longmemory/.

  • Renamed dashboard routes, VS Code command IDs, MCP names, plugin IDs, integration folders, package metadata, assets, and documentation.

  • Moved the canonical repository to https://github.com/CaviraOSS/LongMemory.

  • Added an idempotent branding and file-header migration/checking tool.

  • Added canonical CaviraOSS file headers to every comment-capable active file, with explicit exceptions for strict JSON, binary assets, generated metadata, license texts, and host-owned byte-exact files.

  • Licensed the repository and primary packages under Apache License 2.0.

  • Retained MIT only for the separately published n8n community package because n8n's strict validator requires it.

Architecture: pre-phase HSG to immutable Hydrograph
Before the rewrite

The archived pre-phase implementation contained separate Python and JavaScript SDKs, HSG memory sectors, direct framework adapters, multiple storage backends, and a permissive five-tool MCP server. Important behavior was spread across package-specific implementations and asynchronous wrappers.

Observed pre-phase limitations included:

  • Separate Python and Node engine paths with behavior drift.

  • Query-time cache and lifecycle mutation.

  • Wall-clock-sensitive ranking and decay behavior.

  • Unbatched or per-document embedding requests in several paths.

  • Direct framework adapters with incomplete asynchronous lifecycle handling.

  • Model-supplied user identifiers in MCP calls instead of server-bound identity.

  • Destructive MCP operations without the current deny-by-default allowlist.

  • Incomplete session isolation in conversational integrations.

  • Retrieval that could return derived claims without the original source text.

  • Dense additive graph scoring that diluted strong lexical, vector, or entity evidence when applied universally.

After the rewrite

  • Replaced the HSG implementation with one immutable Hydrograph substrate.

  • Made memory content, vectors, content hashes, provenance, valid-time history, and identity immutable after ingest.

  • Separated mutable activation and lifecycle envelopes from source truth.

  • Made library, CLI, HTTP, MCP, benchmark, dashboard, and editor surfaces call the same engine instead of maintaining parallel implementations.

  • Standardized all durable storage around an in-memory store or SQLite.

  • Separated recorded time from valid time throughout ingest, recall, history, supersession, and conflict handling.

  • Added executable typed edges, world hierarchy, entities, grounding, contracts, conflict reporting, and explainable ingest/recall traces.

  • Bound tenant, user, project, team, role, agent, task, and framework identity at trusted runtime boundaries rather than accepting model impersonation.

  • Kept recall finite, read-only, permission-aware, and token bounded.

Rewrite phase history

The new system was developed in explicit capability phases. Each phase extended the same shared substrate instead of creating a new parallel engine.

Foundation and temporal substrate

  • Phase 1 - package foundation: established the TypeScript package, public entry points, shared engine rule, CLI/server shells, and invariant documents.

  • Phase 2 - memory substrate: added immutable nodes, content identity, facets, state envelopes, provenance, working memory, and core stores.

  • Phase 3 - temporal model: added recorded time, observed time, valid-time intervals, point-in-time filtering, supersession, and timeline reconstruction.

  • Phase 4 - executable edges: introduced typed edge handlers for support, contradiction, containment, derivation, grounding, equivalence, supersession, and semantic shift.

  • Phase 5 - entities: added conservative entity resolution, aliases, candidate indexing, merge/create decisions, and temporal entity identity.

  • Phase 6 - worlds: introduced endocortex/exocortex worlds, parent-child hierarchy, zones, containment, and scoped retrieval.

  • Phase 7 - grounding: added source-backed facts, reliability, freshness, grounding thresholds, and world-grounded evidence gates.

  • Phase 8 - contracts: added memory contracts, required/forbidden evidence, contradiction policy, confidence rules, and validation boundaries.

  • Phase 9 - mathematical primitives: added activation, salience, decay, Hopfield/Oja-style updates, sketches, similarity, and scoring formulas.

Recall and memory evolution

  • Phase 10 - strict recall: combined temporal validity, confidence, contracts, contradictions, grounding, entities, and bounded ranking.

  • Phase 11 - historical recall: preserved superseded facts and enabled point-in-time reconstruction without mutating current truth.

  • Phase 12 - world-grounded recall: required current external evidence for answers that depend on live world state.

  • Phase 13 - associative recall: added BM25, semantic vectors, entities, activation, graph spread, session signals, diversity, and context assembly.

  • Phase 14 - reconsolidation: added explicit, immutable memory revision and supersession rather than in-place content rewriting.

  • Phase 15 - consolidation: added bounded combination of related memories, provenance preservation, and opt-in maintenance.

  • Phase 16 - compression: added deterministic summaries and layered context compression while preserving lossless source text.

  • Phase 17 - ingest pipeline: implemented the production staged ingest path, extraction, identity, graph updates, entities, temporal state, and traces.

Persistence and public surfaces

  • Phase 18 - SQLite: added durable nodes, edges, worlds, entities, grounded facts, sketches, lifecycle state, migrations, integrity, and atomic updates.

  • Phase 19 - public API: stabilized createMemory, ingest, recall, explain, world/entity access, timeline, stats, decay, reinforcement, and shutdown.

  • Phase 20 - HTTP server: added dependency-light authenticated API routes, response envelopes, validation, rate limiting, concurrency limits, CORS, telemetry, health, and Server-Timing.

  • Phase 21 - CLI: added stable JSON automation, interactive TUI behavior, project scoping, ingest/recall/maintenance commands, and server management.

  • Phase 22 - legacy migration: added SQLite/JSON/JSONL import, duplicate and corruption handling, relation restoration, integrity checks, and reports.

Intelligence, projects, and interoperability

  • Phase 23 - connector intelligence: added normalized source documents, permissions, provenance, cursors, rate limits, and import plans.

  • Phase 24 - connector catalog: added GitHub, local files, Markdown, web, feeds, cloud documents, and extensible transport/mapper registries.

  • Phase 25 - project memory: added tenant/organization/project worlds, architecture, decisions, tasks, conventions, failures, handoffs, and code impact tied to persisted source snapshots.

  • Phase 26 - multilingual memory: added script/language detection, code-switch handling, multilingual tokenization, transliteration, cross-script aliases, and optional provenance-marked translation.

  • Phase 27 - MCP: added stdio and authenticated Streamable HTTP transports, 13 governed tools, readable resources, prompts, audit logging, read-only mode, and a deny-by-default tool allowlist.

  • Phase 28 - recovered features: selectively restored useful archived behavior without restoring query-time mutation or duplicate engines.

  • Phase 29 - environment embeddings: added OpenAI-compatible, Gemini, AWS Bedrock, Ollama, Siray, local HTTP, and deterministic fallback stacks with dimensions, tiers, retries, timeouts, and batched embedding support.

  • Phase 30 - performance hardening: added bounded ingest/recall checks, cached immutable preparation, entity indexes, and large-world regression gates.

  • Phase 31 - retrieval quality: added lossless evidence text, targeted matrix retrieval, calibrated feature fusion, aspect-aware evidence selection, polarity/exception coverage, and deterministic retrieval diagnostics.

Retrieval redesign and measured outcomes

  • Split speaker identity from ordinary body terms so common participant names no longer dominated BM25 ranking.

  • Unified query and document token analysis to remove asymmetric stemming and tokenization behavior.

  • Added lightweight stemming and lexical length calibration.

  • Disabled RM3 expansion by default after controlled runs showed query drift and lower evidence quality when the initial top results were wrong.

  • Added a bounded lexical coverage/phrase reranker for shortlist refinement.

  • Replaced quadratic repeated MMR similarity scans with incremental cached maximum-similarity tracking.

  • Added lossless evidence reconstruction so answers receive original source text alongside derived facets and claims.

  • Added matrix-assisted retrieval for universal, contrast, and exception-style questions. Kept it targeted after global application reduced recall.

  • Preserved the legacy linear calibrated signal blend after RRF experiments degraded Hit@K and MRR by discarding useful score magnitude.

  • Reduced recall hot-path CPU by adding ASCII script fast paths, eliminating repeated script detection, hoisting tokenizer expressions, avoiding token object materialization, and caching node token counts.

  • In measured LoCoMo profiling, recall compute at k=20 fell from roughly 147 ms to roughly 89 ms, and ingest CPU fell from 14-46 ms per node to about 10 ms per node without changing retrieval outputs in the controlled check.

  • The complete official comparison made LongMemory the only evaluated provider to complete all selected LongMemEval and LoCoMo cases in that run; competitor failures and incomplete denominators remained explicit rather than being rendered as zero scores.

Deterministic memory lifecycle

  • Replaced destructive/archive-style decay with deterministic envelope-only decay.

  • Added hot, warm, and cold base rates adjusted by salience, confidence, grounding, retention, procedural/emotional value, contradiction, source support, and reinforcement.

  • Made associative recall project decay without persisting it.

  • Added explicit cursor-bounded runDecay() maintenance for persistence.

  • Added monotonic, diminishing explicit reinforcement.

  • Guaranteed that decay and reinforcement never rewrite content, summaries, vectors, provenance, IDs, hashes, or temporal truth.

  • Added SQLite atomicity and audit records for persisted maintenance.

Governed project and agent memory

  • Added project world hierarchies and project-isolated connector sync, recall, timelines, decisions, tasks, failures, and context handoffs.

  • Added immutable reusable Skills with versions, triggers, instructions, validation, resources, archives, and agent bindings.

  • Added snapshot-derived CodeGraph symbol search, callers, callees, and reverse impact paths.

  • Added provider-neutral past-agent session import with original role, order, timestamp, tool-call, provider, and source provenance.

  • Added governed Chat Memory, Skill, LLM-Wiki, and CodeGraph assets.

  • Added candidate/approved/deprecated/archived/failed lifecycle states.

  • Added private/project/team/restricted/agent/task visibility and deny-first ACLs for user, team, role, agent, task, and framework identities.

  • Added deterministic token-budgeted loadouts with direct, summary, tool, and reference injection modes.

  • Added portable manifests with MCP discovery URIs and optional A2A-compatible Agent Card metadata without claiming a complete A2A task server.

Session porter and AI Wiki

  • Added read-only discovery and import adapters for Claude Code, Codex, OpenCode, Gemini CLI, VS Code Copilot Chat, Cline, and raw DeepSeek Harness records.

  • Added portable session IR, preview, select, port, verify, and interactive TUI workflows.

  • Made import idempotency depend on source harness plus native session ID and an immutable revision hash.

  • Made changed sessions create new immutable revisions; unchanged sessions skip.

  • Isolated per-session failures and included them in structured summaries.

  • Added explicit --jsonl progress while preserving one-document JSON for ordinary automation.

  • Added deterministic conversation-to-Markdown LLM-Wiki conversion with source provenance and optional agent binding.

  • Explicitly rejected unsupported compressed DeepSeek records rather than silently parsing partial concatenated Zstandard frames.

CLI and developer experience

  • Added a native terminal application with Library, Review, and Transfer views.

  • Added init, status, memory list, ingest --stdin, recall, explain, timeline, worlds, entities, migration, maintenance, project, Skill, asset, CodeGraph, agent, session, connector, and benchmark commands.

  • Added stable non-interactive JSON and optional JSONL progress.

  • Added workspace-aware .longmemory/project.db defaults and global overrides for database, project, user, and working directory.

  • Canonicalized executable and module paths so linked npm shims work on Windows, macOS, and Linux.

  • Kept prompting exclusive to explicitly interactive workflows.

HTTP API and operational controls

  • Added /health, /v1/ingest, /v1/recall, /v1/explain/:id, /v1/worlds, /v1/worlds/:id, /v1/entities/:id, /v1/timeline, /v1/stats, and /v1/runtime.

  • Added consistent {data, meta} success envelopes and structured errors.

  • Added optional bearer or X-API-Key authentication.

  • Added maximum payload, active request, fixed-window rate limit, CORS, auth logging, runtime telemetry, and graceful close controls.

  • Added platform-assigned PORT support while preserving local port 7331.

  • Added server-side-only dashboard credentials and removed client-bundled API secrets.

MCP and integrations

  • Added local stdio and authenticated Streamable HTTP MCP transports.

  • Added 13 high-level tools for context, recall, ingest, decisions, tasks, explanation, conflicts, connectors, Skills, CodeGraph, and governed assets.

  • Added project resources, prompts, tool filtering, read-only mode, and JSONL audit logs.

  • Removed the experimental assimilation protocol and all duplicate integration engine wrappers.

  • Added a strict n8n community node with Recall, Store, Explain, and Stats operations and usableAsTool support.

  • Added an Agent Plugins 1.0 bundle for OpenClaw and compatible hosts.

  • Added native Claude Code, Codex/ChatGPT desktop, and Gemini CLI packages.

  • Added Cline, Continue, and LibreChat MCP configuration packs.

  • Added runnable native-MCP examples for CrewAI, AutoGen, LangGraph/LangChain, OpenAI Agents SDK, and PydanticAI.

  • Documented native MCP attachment for Dify, Flowise, and other compatible hosts.

Dashboard and VS Code

  • Added a production Next.js dashboard for health, memory browsing, ingest, project selection, activity, search, timelines, decay, settings, and chat.

  • Added a same-origin server proxy so API credentials stay server-side.

  • Removed dashboard controls that attempted to mutate immutable memory content.

  • Added a native VS Code activity view, status indicators, quick note and selection ingest, recall, project context, explanation, reinforcement, decay, and session import.

  • Added explicit AI-change sessions that capture bounded, redacted patches from editor buffers and direct workspace writes.

  • Kept extension-detection heuristics opt-in, low confidence, review-only, and never automatically persisted.

  • Added process tracking and Windows process-tree shutdown for CLI subprocesses.

  • Packaged the extension as longmemory-vscode with its Apache license.

Connectors and documents

  • Added a unified connector contract with inspection, preview, dry-run, permissions, provenance, source versions, cursors, rate limits, updates, and deletions.

  • Added repository and local filesystem intelligence, GitHub transport, Markdown, websites, feeds, cloud services, communication tools, project systems, and data connectors.

  • Added extraction for PDF, DOCX, HTML, Markdown, plain text, source code, audio transcripts, and video-derived audio.

  • Added atomic import plans instead of allowing connectors to write arbitrary memory chunks directly.

Embedding providers

  • Added OpenAI and OpenAI-compatible endpoints.

  • Added Gemini Embedding 001 with required truncation renormalization.

  • Added AWS Bedrock Titan, Ollama, Siray, local HTTP, OpenRouter-compatible, and DashScope-compatible configurations.

  • Added environment-driven fast, smart, deep, and hybrid tiers.

  • Added timeout, retry, dimension, model, batch size, and ordered fallback controls.

  • Added batched Gemini embedding requests and provider-specific benchmark batch sizes.

  • Made official semantic benchmark runs fail closed when a fallback provider was actually used.

Benchmark and release engineering

  • Replaced the archived benchmark scripts with a manifest-driven TypeScript runner and deterministic checkpoint/resume model.

  • Added LongMemEval, LoCoMo, BEAM-1M, BEAM-10M, and deterministic smoke datasets.

  • Added answer-at-every-cutoff evaluation with distinct answerer and judge roles.

  • Added evidence Hit@K, recall, rank-weighted precision, MRR, nDCG, answer accuracy, temporal categories, abstention, stale-evidence, latency, context, and optional cost metrics.

  • Added explicit incomplete-run accounting and N/A values instead of silently converting missing evidence or judgments to zero.

  • Added comparative adapters for Mem0, Zep/Graphiti, Cognee, and Supermemory.

  • Added a stable public K=5 scorecard and benchmark release gates.

  • Added clean npm package builds that remove stale dist files before emission.

  • Added package-content inspection to prevent test, legacy, or removed artifacts from reaching npm.

  • Added Docker, Compose, Heroku, Railway, Render, DigitalOcean, and Vercel dashboard deployment files.

  • Added npm, n8n, VS Code, container, and trusted-publisher workflow definitions.

  • Added dependency auditing and a zero-vulnerability dashboard release lockfile.

  • Added release validation without shipping active test suites or Vitest.

Breaking changes from pre-phase OpenMemory

  • The product and repository are now LongMemory.

  • The primary JavaScript package changed from openmemory-js to longmemory.

  • The primary CLI changed from opm/openmemory-js to longmemory.

  • Environment variables changed from OM_*/OPENMEMORY_* to LONGMEMORY_*, with only selected internal configuration aliases retained.

  • The default network port changed from 8080 to 7331.

  • The default workspace database moved to .longmemory/project.db.

  • The former embedded Python engine was removed. Python now accesses the single TypeScript engine through the self-hosted HTTP API.

  • Postgres, Valkey, Qdrant, and separate Python-engine storage claims from the archived line are not part of the LongMemory 1.0 public contract.

  • Direct LangChain/CrewAI/AutoGen memory wrappers were replaced by official MCP client surfaces and thin examples.

  • The former permissive MCP query/store/get/delete/list interface was replaced by governed high-level tools with server-bound identity and audit policy.

  • Destructive delete and model-selected identity are not exposed through the new MCP surface.

  • Query-time mutation, implicit wall-clock decay, and hidden maintenance timers were removed.

  • Compatibility is delivered through temporary registry bridge packages, not by preserving old runtime namespaces throughout the engine.

Migration guidance

JavaScript:

npm uninstall openmemory-js
npm install longmemory

Python:

pip uninstall openmemory-py
pip install longmemory-sdk

Data:

longmemory migrate \
  --from ./legacy.db \
  --to ./longmemory.db \
  --report ./migration-report.json

See MIGRATION.md and docs/migration.md for package, database, and session migration details.

Archived pre-phase lineage

The following entries describe the archived OpenMemory implementation and are not statements about the LongMemory 1.0 public contract.

[OpenMemory 1.3.0] - 2025-12-20

Added

  • Simplified the archived Python SDK to a zero-configuration Memory() API.

  • Reworked the archived benchmark suite in TypeScript with LongMemEval and multi-backend comparison.

  • Added Memory.wipe() and database reset support for benchmark isolation.

  • Added embedding model, vector dimension, database path, and provider environment overrides.

  • Added additional Postgres vector-store operation logging.

Fixed

  • Corrected Ollama model override and embedding dimension handling.

  • Corrected deep-tier semantic embedding configuration.

  • Reduced benchmark cross-user contamination through reset handling.

  • Corrected benchmark environment loading and dataset path resolution.

Changed

  • Replaced the former OpenMemory Python class entry point with Memory.

  • Moved the benchmark implementation from Python to TypeScript.

  • Expanded archived Python and JavaScript SDK documentation.

[OpenMemory 1.2.2] - 2024-11

  • Fixed consolidation edge cases.

  • Improved multi-user query isolation.

  • Corrected vector dimension handling.

[OpenMemory 1.2.1] - 2024-11

  • Improved large-dataset performance.

  • Improved archived sector classification behavior.

[OpenMemory 1.2.0] - 2024-10

  • Added the original multi-sector memory architecture.

  • Added the original cognitive decay system.

  • Added reflection and consolidation.

[OpenMemory 1.1.0] - 2024-09

  • Added the initial TypeScript SDK.

  • Added SQLite vector storage.

  • Added basic query and add operations.

[OpenMemory 1.0.0] - 2024-08

  • Published the original Python SDK.

  • Established the initial local-first architecture.

  • Added basic memory operations.

512 files · 459 not served here

Advantages

  • Supports temporal truth by separating recorded time from valid time, allowing for historical accuracy. evidence: README.md
  • Provides immutable memory where content and provenance are preserved, preventing accidental overwrites. evidence: README.md
  • Enforces governance policies for projects, teams, and roles to ensure secure and scoped memory access. evidence: README.md

Limitations

  • Requires manual configuration for network deployment, including API key management and TLS termination. evidence: README.md
  • The Python SDK requires a running LongMemory server instance for persistent storage. evidence: README.md

Security

Risk level: Low risk

This brain runs

Install scripts (1)

  • Makefile (make install): corepack enable; pnpm install --frozen-lockfile

MCP servers (2)

  • integrations/claude-code-longmemory/.mcp.json (longmemory): longmemory mcp --project current
  • integrations/gemini-cli-longmemory/gemini-extension.json (longmemory): longmemory mcp --project current
  • Low risk: scripts-present (benchmarks/src/official.ts)
  • Low risk: scripts-present (src/cli/index.ts)
  • Low risk: scripts-present (tools/check-integrations.mjs)

Attribution and licence

Created by
CaviraOSS
First committed by
@nullure
Declared authors
  • CaviraOSS (declared in package.json)
Publisher
Cavira (Community)
Source
CaviraOSS/LongMemory · commit aec98ce
Licence
Apache License 2.0 (files served here)
  • ATTRIBUTION.md (written by AutoGineer)
Synced

AutoGineer is not affiliated with or endorsed by CaviraOSS.

Install from source

459 of its files are not served here, so there is no ZIP. CaviraOSS/LongMemory at aec98ce