๐ฌTechnical Deep Dive
Full Specifications [+]
Quick Commands
git clone https://github.com/zilliztech/memsearch pip install memsearch โ๏ธ Free2AITools Nexus Index V2.0
๐ฌ Index Insight
FNI V2.0 for memsearch: Semantic (S:50), Authority (A:0), Popularity (P:67), Recency (R:100), Quality (Q:50).
Verification Authority
๐ Specs
- Language
- Python
- License
- MIT
- Version
- 1.0.0
Usage documentation not yet indexed for this tool.
๐ View Source Code โTechnical Documentation
memsearch
Cross-platform semantic memory for AI coding agents.
Why memsearch?
- ๐ All Platforms, One Memory โ memories flow across Claude Code, OpenClaw, OpenCode, and Codex CLI. A conversation in one agent becomes searchable context in all others โ no extra setup
- ๐ฅ For Agent Users, install a plugin and get persistent memory with zero effort; for Agent Developers, use the full CLI and Python API to build memory and harness engineering into your own agents
- ๐ Markdown is the source of truth โ inspired by OpenClaw. Your memories are just
.mdfiles โ human-readable, editable, version-controllable. Milvus is a "shadow index": a derived, rebuildable cache - ๐ Progressive retrieval, hybrid search, smart dedup, live sync โ 3-layer recall (search โ expand โ transcript); dense vector + BM25 sparse + RRF reranking; SHA-256 content hashing skips unchanged content; file watcher auto-indexes in real time
๐งโ๐ป For Agent Users
Pick your platform, install the plugin, and you're done. Each plugin captures conversations automatically and provides semantic recall with zero configuration.
For Claude Code Users
# Install
/plugin marketplace add zilliztech/memsearch
/plugin install memsearch
# Restart Claude Code to activate the pluginAfter restarting, just chat with Claude Code as usual. The plugin captures every conversation turn automatically.
Verify it's working โ after a few conversations, check your memory files:
ls .memsearch/memory/ # you should see daily .md files
cat .memsearch/memory/$(date +%Y-%m-%d).mdRecall memories โ two ways to trigger:
/memory-recall what did we discuss about Redis?Or just ask naturally โ Claude auto-invokes the skill when it senses the question needs history:
We discussed Redis caching before, what was the TTL we chose?For Codex CLI Users
# Install
git clone --depth 1 https://github.com/zilliztech/memsearch.git
bash memsearch/plugins/codex/scripts/install.sh
codex --yolo # needed for ONNX model network accessAfter installing, chat as usual. Hooks capture and summarize each turn.
Verify it's working:
ls .memsearch/memory/Recall memories โ use the skill:
$memory-recall what did we discuss about deployment?For OpenClaw Users
# Install from ClawHub
openclaw plugins install --force clawhub:memsearch
openclaw config set plugins.entries.memsearch.hooks.allowConversationAccess true
openclaw config set plugins.entries.memsearch.hooks.allowPromptInjection true
openclaw gateway restartAfter installing, chat in TUI as usual. The plugin captures each turn automatically.
Verify it's working โ memory files are stored in your agent's workspace:
# For the main agent:
ls ~/.openclaw/workspace/.memsearch/memory/
# For other agents (e.g. work):
ls ~/.openclaw/workspace-work/.memsearch/memory/Recall memories โ two ways to trigger:
/memory-recall what was the batch size limit we set?Or just ask naturally โ the LLM auto-invokes memory tools when it senses the question needs history:
We discussed batch size limits before, what did we decide?For OpenCode Users
// In ~/.config/opencode/opencode.json
{ "plugin": ["@zilliz/memsearch-opencode"] }After installing, chat in TUI as usual. A background daemon captures conversations.
Verify it's working:
ls .memsearch/memory/ # daily .md files appear after a few conversationsRecall memories โ two ways to trigger:
/memory-recall what did we discuss about authentication?Or just ask naturally โ the LLM auto-invokes memory tools when it senses the question needs history:
We discussed the authentication flow before, what was the approach?๐ OpenCode Plugin docs
โ๏ธ Configuration (all platforms)
All plugins share the same memsearch backend. Configure once, works everywhere.
Embedding
Defaults to ONNX bge-m3 โ runs locally on CPU, no API key, no cost. On first launch the model (~558 MB) is downloaded from HuggingFace Hub.
memsearch config set embedding.provider onnx # default โ local, free
memsearch config set embedding.provider openai # needs OPENAI_API_KEY
memsearch config set embedding.provider ollama # local, any modelAll providers and models: Configuration โ Embedding Provider
Milvus Backend
Just change milvus_uri (and optionally milvus_token) to switch between deployment modes:
Milvus Lite (default) โ zero config, single file. Great for getting started:
# Works out of the box, no setup needed
memsearch config get milvus.uri # โ ~/.memsearch/milvus.dbโญ Zilliz Cloud (recommended) โ fully managed, free tier available โ sign up ๐:
memsearch config set milvus.uri "https://in03-xxx.api.gcp-us-west1.zillizcloud.com"
memsearch config set milvus.token "your-api-key"โญ Sign up for a free Zilliz Cloud cluster
You can sign up on Zilliz Cloud to get a free cluster and API key.
Self-hosted Milvus Server (Docker) โ for advanced users
For multi-user or team environments with a dedicated Milvus instance. Requires Docker. See the official installation guide.
memsearch config set milvus.uri http://localhost:19530๐ Full configuration guide: Configuration ยท Platform comparison
Capture Summarization Routing
Each plugin keeps its native capture summarizer unless you override it explicitly:
memsearch config set plugins.codex.summarize.model gpt-5.1-codex-mini
memsearch config set plugins.opencode.summarize.model anthropic/claude-haikuAdvanced users can route plugin summarization through a memsearch-managed API provider:
memsearch config set llm.providers.openai.type openai
memsearch config set llm.providers.openai.model gpt-5-mini
memsearch config set llm.providers.openai.api_key env:OPENAI_API_KEY
memsearch config set plugins.codex.summarize.provider openaiLeave plugins.<platform>.summarize.provider empty or set it to native to preserve the default behavior. Plugin-specific summarize settings do not fall back to llm.model.
You can also disable automatic capture for a project while keeping the plugin installed:
memsearch config set plugins.codex.summarize.enabled false --projectAdvanced Memory Maintenance
Plugins can optionally maintain higher-level project and user notes in the background. These tasks are disabled by default and run only when a plugin wakes them after a session/turn, the journal input changed, and min_interval_hours has elapsed.
memsearch config set plugins.codex.project_review.enabled true --project
memsearch config set plugins.codex.project_review.provider native --project
memsearch config set plugins.codex.project_review.min_interval_hours 24 --project
memsearch config set plugins.codex.project_review.output_file .memsearch/PROJECT.md --project
memsearch config set plugins.codex.user_profile.enabled true --project
memsearch config set plugins.codex.user_profile.output_file .memsearch/USER.md --projectproject_review summarizes durable project state such as active threads, decisions, risks, and next steps. user_profile captures reusable user preferences, working style, recurring goals, and background context. Both read .memsearch/memory by default; set input_dir if your journal files live somewhere else.
Use provider = "native" to reuse the current agent's own non-interactive model path, or point the task at a named [llm.providers.<name>] API provider. Custom prompt files can be configured with prompts.project_review and prompts.user_profile.
The memory-config skill, installed with the plugins, can inspect the current setup, explain these options, and make safe project-scoped changes from natural-language requests.
What can you use it for?
- Resume debugging threads โ ask how a similar Redis, Docker, database, or deployment issue was fixed last time.
- Recover decision rationale โ find why the project chose one architecture, library, migration path, or API design over another.
- Trace feature history โ understand how a feature evolved across sessions, including the files changed and tradeoffs discussed.
- Do code archaeology โ ask when and why a module, config, or workflow was changed before touching it again.
- Find the right session to resume โ ask which previous conversation covered a topic, recover the relevant context, and continue from there.
- Carry context across agents โ keep Claude Code, Codex CLI, OpenClaw, and OpenCode working from the same project memory.
๐ ๏ธ For Agent Developers
Beyond ready-to-use plugins, memsearch provides a complete CLI and Python API for building memory into your own agents. Whether you're adding persistent context to a custom agent, building a memory-augmented RAG pipeline, or doing harness engineering โ the same core engine that powers the plugins is available as a library.
๐๏ธ Architecture Overview
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
โ ๐งโ๐ป For Agent Users (Plugins) โ
โ โโโโโโโโโโโโ โโโโโโโโโโโโ โโโโโโโโโโโโ โโโโโโโโโโ โโโโโโโโ โ
โ โ Claude โ โ OpenClaw โ โ OpenCode โ โ Codex โ โ Your โ โ
โ โ Code โ โ Plugin โ โ Plugin โ โ Plugin โ โ App โ โ
โ โโโโโโฌโโโโโโ โโโโโโฌโโโโโโ โโโโโโฌโโโโโโ โโโโโฌโโโโโ โโโโฌโโโโ โ
โ โโโโโโโโโโโโโโโดโโโโโโโโโโโโโดโโโโโโโโโโโโดโโโโโโโโโ โ
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฌโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค
โ ๐ ๏ธ For Agent Developers โ Build your own with โ โ
โ โโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โ
โ โ memsearch CLI / Python API โ โ
โ โ index ยท search ยท expand ยท watch ยท compact โ โ
โ โโโโโโโโโโโโโโโโโโโโโโโโโโโฌโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โ
โ โโโโโโโโโโโโโโโโโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โ
โ โ Core: Chunker โ Embedder โ Milvus โ โ
โ โ Hybrid Search (BM25 + Dense + RRF) โ โ
โ โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โ
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค
โ ๐ Markdown Files (Source of Truth) โ
โ memory/2026-03-27.md ยท memory/2026-03-26.md ยท ... โ
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโPlugins sit on top of the CLI/API layer. The API handles indexing, searching, and Milvus sync. Markdown files are always the source of truth โ Milvus is a rebuildable shadow index. Everything below the plugin layer is what you use as an agent developer.
How Plugins Work (Claude Code as example)
Capture โ after each conversation turn:
User asks question โ Agent responds โ Stop hook fires
โ
โโโโโโโโโโโโโโโโโโโโโโ
โผ
Parse last turn
โ
โผ
LLM summarizes (haiku)
"- User asked about X."
"- Claude did Y."
โ
โผ
Append to memory/2026-03-27.md
with anchor
โ
โผ
memsearch index โ MilvusRecall โ 3-layer progressive search:
User: "What did we discuss about batch size?"
โ
โผ
L1 memsearch search "batch size" โ ranked chunks
โ (need more?)
โผ
L2 memsearch expand โ full .md section
โ (need original?)
โผ
L3 parse-transcript โ raw dialogue ๐ Markdown as Source of Truth
Plugins append โโโ .md files โโโ human editable
โ
โผ
memsearch watch (live watcher)
โ
detects file change
โ
โผ
re-chunk changed .md
โ
hash each chunk (SHA-256)
โ
โโโโโโโโโโโโโดโโโโโโโโโโโโ
โผ โผ
hash unchanged? hash is new/changed?
โ skip (no API call) โ embed โ upsert to Milvus
โ โ
โโโโโโโโโโโโโฌโโโโโโโโโโโโ
โผ
โโโโโโโโโโโโโโโโโโโโ
โ Milvus (shadow) โ
โ always in sync โ
โ rebuildable โ
โโโโโโโโโโโโโโโโโโโโ๐ฆ Installation
# Install as a global CLI tool โ recommended when you mainly use the
# `memsearch` command or any of the agent plugins (Claude Code, Codex,
# OpenClaw, OpenCode), which all shell out to the CLI.
uv tool install memsearch # via uv
pipx install memsearch # via pipx
pip install memsearch # plain pip
# Install as a project dependency โ use this if you want to import
# `memsearch` from your own Python code (e.g. via the MemSearch class).
uv add memsearch # via uv, adds to pyproject.toml
pip install memsearch # into an activated venvOptional embedding providers
# As a CLI tool (recommended โ local ONNX, no API key)
uv tool install "memsearch[onnx]"
pipx install "memsearch[onnx]"
pip install "memsearch[onnx]"
# As a project dependency
uv add "memsearch[onnx]"
# Other options: [openai], [google], [voyage], [jina], [mistral], [ollama], [local], [all]๐ Python API โ Give Your Agent Memory
from memsearch import MemSearch
mem = MemSearch(paths=["./memory"])
await mem.index() # index markdown files
results = await mem.search("Redis config", top_k=3) # semantic search
scoped = await mem.search("pricing", top_k=3, source_prefix="./memory/product")
print(results[0]["content"], results[0]["score"]) # content + similarityFull example โ agent with memory (OpenAI) โ click to expand
import asyncio
from datetime import date
from pathlib import Path
from openai import OpenAI
from memsearch import MemSearch
MEMORY_DIR = "./memory"
llm = OpenAI() # your LLM client
mem = MemSearch(paths=[MEMORY_DIR]) # memsearch handles the rest
def save_memory(content: str):
"""Append a note to today's memory log (OpenClaw-style daily markdown)."""
p = Path(MEMORY_DIR) / f"{date.today()}.md"
p.parent.mkdir(parents=True, exist_ok=True)
with open(p, "a") as f:
f.write(f"\n{content}\n")
async def agent_chat(user_input: str) -> str:
# 1. Recall โ search past memories for relevant context
memories = await mem.search(user_input, top_k=3)
context = "\n".join(f"- {m['content'][:200]}" for m in memories)
# 2. Think โ call LLM with memory context
resp = llm.chat.completions.create(
model="gpt-5-mini",
messages=[
{"role": "system", "content": f"You have these memories:\n{context}"},
{"role": "user", "content": user_input},
],
)
answer = resp.choices[0].message.content
# 3. Remember โ save this exchange and index it
save_memory(f"## {user_input}\n{answer}")
await mem.index()
return answer
async def main():
# Seed some knowledge
save_memory("## Team\n- Alice: frontend lead\n- Bob: backend lead")
save_memory("## Decision\nWe chose Redis for caching over Memcached.")
await mem.index() # or mem.watch() to auto-index in the background
# Agent can now recall those memories
print(await agent_chat("Who is our frontend lead?"))
print(await agent_chat("What caching solution did we pick?"))
asyncio.run(main())Anthropic Claude example โ click to expand
pip install memsearch anthropicimport asyncio
from datetime import date
from pathlib import Path
from anthropic import Anthropic
from memsearch import MemSearch
MEMORY_DIR = "./memory"
llm = Anthropic()
mem = MemSearch(paths=[MEMORY_DIR])
def save_memory(content: str):
p = Path(MEMORY_DIR) / f"{date.today()}.md"
p.parent.mkdir(parents=True, exist_ok=True)
with open(p, "a") as f:
f.write(f"\n{content}\n")
async def agent_chat(user_input: str) -> str:
# 1. Recall
memories = await mem.search(user_input, top_k=3)
context = "\n".join(f"- {m['content'][:200]}" for m in memories)
# 2. Think โ call Claude with memory context
resp = llm.messages.create(
model="claude-sonnet-4-6",
max_tokens=1024,
system=f"You have these memories:\n{context}",
messages=[{"role": "user", "content": user_input}],
)
answer = resp.content[0].text
# 3. Remember
save_memory(f"## {user_input}\n{answer}")
await mem.index()
return answer
async def main():
save_memory("## Team\n- Alice: frontend lead\n- Bob: backend lead")
await mem.index()
print(await agent_chat("Who is our frontend lead?"))
asyncio.run(main())Ollama (fully local, no API key) โ click to expand
pip install "memsearch[ollama]"
ollama pull nomic-embed-text # embedding model
ollama pull llama3.2 # chat modelimport asyncio
from datetime import date
from pathlib import Path
from ollama import chat
from memsearch import MemSearch
MEMORY_DIR = "./memory"
mem = MemSearch(paths=[MEMORY_DIR], embedding_provider="ollama")
def save_memory(content: str):
p = Path(MEMORY_DIR) / f"{date.today()}.md"
p.parent.mkdir(parents=True, exist_ok=True)
with open(p, "a") as f:
f.write(f"\n{content}\n")
async def agent_chat(user_input: str) -> str:
# 1. Recall
memories = await mem.search(user_input, top_k=3)
context = "\n".join(f"- {m['content'][:200]}" for m in memories)
# 2. Think โ call Ollama locally
resp = chat(
model="llama3.2",
messages=[
{"role": "system", "content": f"You have these memories:\n{context}"},
{"role": "user", "content": user_input},
],
)
answer = resp.message.content
# 3. Remember
save_memory(f"## {user_input}\n{answer}")
await mem.index()
return answer
async def main():
save_memory("## Team\n- Alice: frontend lead\n- Bob: backend lead")
await mem.index()
print(await agent_chat("Who is our frontend lead?"))
asyncio.run(main())๐ Full Python API reference: Python API docs
โจ๏ธ CLI Usage
Setup:
memsearch config init # interactive setup wizard
memsearch config set embedding.provider onnx # switch embedding provider
memsearch config set milvus.uri http://localhost:19530 # switch Milvus backendIndex & Search:
memsearch index ./memory/ # index markdown files
memsearch index ./memory/ ./notes/ --force # re-embed everything
memsearch search "Redis caching" # hybrid search (BM25 + vector)
memsearch search "auth flow" --top-k 10 --json-output # JSON for scripting
memsearch expand # show full section around a chunk Live Sync & Maintenance:
memsearch watch ./memory/ # live file watcher (auto-index on change)
memsearch compact # LLM-powered chunk summarization
memsearch stats # show indexed chunk count
memsearch reset --yes # drop all indexed data and rebuild๐ Full CLI reference with all flags: CLI docs
โ๏ธ Configuration
Embedding and Milvus backend settings โ Configuration (all platforms)
Settings priority: Built-in defaults โ ~/.memsearch/config.toml โ .memsearch.toml โ CLI flags.
๐ Full config guide: Configuration
๐ Links
- ๐ Documentation โ full guides, API reference, and architecture details
- ๐ Platform Plugins โ Claude Code, OpenClaw, OpenCode, Codex CLI
- ๐ก Design Philosophy โ why markdown, why Milvus, competitor comparison
- ๐ฆ OpenClaw โ the memory architecture that inspired memsearch
- ๐๏ธ Milvus | Zilliz Cloud โ the vector database powering memsearch
๐ค Contributing
Bug reports, feature requests, and pull requests are welcome! See the Contributing Guide for development setup, testing, and plugin development instructions. For questions and discussions, join us on Discord.
๐ License
๐ Quick Start
# Install
git clone --depth 1 https://github.com/zilliztech/memsearch.git
bash memsearch/plugins/codex/scripts/install.sh
codex --yolo # needed for ONNX model network accessAI Summary: Based on GitHub metadata. Not a recommendation.
๐ก๏ธ Tool Transparency Report
Technical metadata sourced from upstream repositories.
๐ Identity & Source
- id
- gh-tool--zilliztech--memsearch
- slug
- zilliztech--memsearch
- source
- github
- author
- zilliztech
- license
- MIT
- tags
- agent-memory, claude-code, claude-code-plugin, clawdbot, memory, openclaw, progressive-disclosure, rag, agent, embeddings, milvus, semantic-search, python, ai-agents, harness, hybrid-search, long-term-memory, opencode, reranker, skills, codex, codex-cli
โ๏ธ Technical Specs
- architecture
- null
- params billions
- null
- context length
- null
- pipeline tag
- other
๐ Engagement & Metrics
- downloads
- 0
- stars
- 1,096
- forks
- 174
- github stars
- 1,096
Data indexed from public sources. Updated daily.