Help us improve
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
By memsy-io
Persists and retrieves decisions, facts, and past context across Claude Code sessions with slash commands, a deep-retrieval subagent, and optional session hooks.
npx claudepluginhub memsy-io/memsy --plugin memsyCapture the substance of this conversation as Memsy memories — decisions, fixes, learnings worth surfacing in future sessions
Check Memsy MCP health, identity, active profile, and surface setup hints
Ingest a structured snapshot of the current codebase into Memsy — so search can answer "how is this project laid out" later
Switch the active Memsy profile / org for this session
Store a fact, decision, or note in Memsy so it surfaces in future sessions
Use this skill when the user asks to recall past context, decisions, or what was discussed previously. Trigger phrases include "what did we decide", "remember when", "have we discussed", "context on X", "do we have anything about Y", "look up", "search past conversations", "find that thing about", or any retrieval-intent question about prior work. Calls memsy_search MCP tool with the topic extracted from the user's question.
Use this skill when the user explicitly asks to save, note, or persist a fact, decision, or piece of context for future sessions. Trigger phrases include "remember that", "save this decision", "note that", "let's remember", "store this", "tag this as", "for future reference", or any clear save-intent statement. Calls memsy_ingest MCP tool to persist the substance.
Use this skill when Memsy MCP tools are unavailable or returning errors — symptoms include "tool not found", "memsy_health unavailable", 401/403 auth errors from Memsy, "ECONNREFUSED" or network errors to api.memsy.io, or when the user explicitly asks to "set up memsy", "install memsy", "memsy isn't working", "memsy auth error", or similar. Diagnoses the failure mode and walks the user through the fix without retrying the broken tool blindly.
Triggered ONLY when the user explicitly types the `/memsy` slash command in Claude Code. Do not auto-fire on natural-language phrasing — the `memsy-recall`, `memsy-remember`, and `memsy-setup` skills handle those without a slash.
Share bugs, ideas, or general feedback.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Persistent memory for Claude Code — memories survive across sessions, projects, and machines
Captures user prompts and agent responses to Memory Engine for persistent context across sessions
Memory compression system for Claude Code - persist context across sessions
Multi-model consensus engine integrating OpenAI Codex CLI, Gemini CLI, and Claude CLI for collaborative code review and problem-solving.
Standalone image generation plugin using Nano Banana MCP server. Generates and edits images, icons, diagrams, patterns, and visual assets via Gemini image models. No Gemini CLI dependency required.
Ultra-compressed communication mode. Cuts ~75% of tokens while keeping full technical accuracy by speaking like a caveman.
Persistent memory for AI agents and applications. 88.12% on LoCoMo — the highest accuracy of any memory library at k≤20, at 4× better score-per-token than mem0.
📚 Docs: docs.memsy.io | 🌐 Website: memsy.io
| Language | Package | Install |
|---|---|---|
| Python | memsy-io/sdks/python | pip install memsy |
| Node.js | memsy-io/sdks/node | npm install @memsy-io/memsy |
Python
from memsy import MemsyClient, EventPayload
client = MemsyClient(base_url="https://api.memsy.io", api_key="msy_...")
# Store a memory
client.ingest([EventPayload(
actor_id="user_1",
session_id="session_1",
kind="user_message",
content="I prefer dark mode in all apps",
)])
# Recall it later
results = client.search("user preferences", actor_id="user_1")
for r in results.results:
print(r.content)
Node.js
import { MemsyClient } from "@memsy-io/memsy";
const client = new MemsyClient({ baseUrl: "https://api.memsy.io", apiKey: "msy_..." });
// Store a memory
await client.ingest([{
actorId: "user_1",
sessionId: "session_1",
kind: "user_message",
content: "I prefer dark mode in all apps",
}]);
// Recall it later
const { results } = await client.search("user preferences", { actorId: "user_1" });
results.forEach(r => console.log(r.content));
| Memsy | mem0 | |
|---|---|---|
| LoCoMo Score | 88.12% | 82.7% |
| Tokens per query | ~1,700 | ~7,000 |
| Score / 1K tokens | 51.8 | 13.1 |
Benchmark: full LoCoMo suite (1,540 questions), GPT-4.1 Mini, single pass, no post-processing. Full results →
Coming soon: Claude MCP, OpenAI, LangChain, LlamaIndex.
memsy/
├── sdks/
│ ├── python/ ← Python SDK (pip install memsy)
│ └── node/ ← Node.js SDK (npm install @memsy-io/memsy)
└── connectors/ ← Claude, OpenAI and other connectors (coming soon)
We welcome contributions to any SDK or connector in this repo. See CONTRIBUTING.md for the full guide — repo layout, per-SDK setup, and PR conventions.
For bugs or feature requests, open an issue.
MIT