Skip to content
MCP server

Agentics

Hosted shared memory for AI agents, with a checkable receipt for every write

Install
Pricing
Freemium
Platforms
API, Web
Runs
Remotely
Version
2.1.0
Listing updated
2 days ago

About Agentics

Agentics is a freemium MCP server for knowledge and memory. It runs as a hosted remote server, so there is nothing to install locally.

Agentics is reached at one hosted address, https://agentics.you/mcp, and signs in once through OAuth so you stay the owner of every connected agent. Memory tools such as remember, recall, update_memory and forget_memory let several agents share notes in plain words, while each write produces a receipt. The pricing page says receipts can be checked without an account.

Beyond memory there are controls for the owner: ask_owner sends you a question, my_rules returns your rules and can Pause all, and the receipt tools (log_action, report_outcome, record_action) write the record of what an agent did. A hub lets one agent hand work to another: send_task, inbox, submit_result and get_result move a task between agents once you switch on permission for the pair, and route_task logs which agent it would choose along with each candidate's odds.

The docs give setup steps for Grok, Claude, ChatGPT, Cursor, Hermes, OpenClaw, Claude Code, Codex, Gemini CLI, VS Code and Windsurf. A connection is limited to 120 calls a minute. Plans count actions (receipts written) rather than agents, and recall is never counted.

Install

Run in a terminal

claude mcp add --transport http agentics https://agentics.you/mcp

Add to Settings → Connectors → Add custom connector

https://agentics.you/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "agentics": {
      "url": "https://agentics.you/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "agentics": {
      "type": "http",
      "url": "https://agentics.you/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "agentics": {
      "serverUrl": "https://agentics.you/mcp"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "agentics": {
      "type": "http",
      "url": "https://agentics.you/mcp"
    }
  }
}

Tools this server provides (15)

Taken from the project documentation; the running server may differ.

remember
Saves a memory in plain words.
recall
Searches saved memories.
update_memory
Corrects a wrong memory and marks the old one replaced.
forget_memory
Marks a memory forgotten so recall stops returning it.
ask_owner
Sends the owner a question.
log_action
Writes an action receipt.
report_outcome
Writes an outcome receipt.
record_action
Records an action, optionally with reasoning and lessons.
my_rules
Returns the owner's rules and can Pause all agents.
send_task
Sends a task to another agent.
inbox
Takes tasks waiting for this agent.
submit_result
Finishes a task with a status and summary.
get_result
Reads a task's status and result summary.
route_task
Chooses which agent should take a task and logs the decision.
list_peers
Lists peer agents with their top scores by kind of work.

What it does

Shared agent memory
remember, recall, update_memory and forget_memory let agents save, search, correct and retire memories.
Receipts for writes
Every memory or action write produces a receipt that can be checked without an account.
Owner questions and rules
ask_owner sends you a question and my_rules returns your rules, including a Pause all that only you can resume.
Task hand-off hub
send_task, inbox, submit_result and get_result pass work between two agents on the same account.
Routing decisions
route_task picks which agent should take a task and logs every candidate's odds without sending it.
Many clients supported
One address works across Grok, Claude, ChatGPT, Cursor, Codex, Gemini CLI, VS Code and others.

Good for

  • Give several coding agents one shared, searchable memory
  • Keep an auditable record of what each agent decided or did
  • Hand a drafting task from one agent to another and read back the result
  • Correct or forget a wrong memory so agents stop recalling it
  • Ask the human owner for approval from inside an agent run

Pricing

Freemium, from $29 a month

Free plan; Pro $29/mo, Team $199/mo, Enterprise from $1,500/mo

Prices change. Check agentics.you before you buy.

Agentics FAQ

What is the server address?

https://agentics.you/mcp over streamable HTTP, with OAuth sign-in.

Why can't agent A send a task to agent B?

send_task returns not_allowed until you turn on the switch for that peer in the console, and paused if Pause all is on.

How many calls can a connection make?

The docs state 120 calls per minute on each connection's token.

Is recall counted against my plan?

No. Plans count receipts written; reads, recall and checking a receipt are never counted.

What does it cost?

Free is $0 with 10,000 stored memories and 10,000 actions a month; Pro is $29 a month, Team $199 and Enterprise starts at $1,500.

Is Agentics free?

Agentics has a free plan, with paid plans for more usage or features. Paid plans start at $29 a month.

How do I add Agentics to Claude Code?

Run this command in your terminal: claude mcp add --transport http agentics https://agentics.you/mcp

What are the best alternatives to Agentics?

The closest alternatives to Agentics are Codebase Memory, XERJ, Claude Code Ultimate Guide, Nautilus Compass Memory and note1.

Alternatives to Agentics

  1. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  2. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  3. MCP server for searching the Claude Code Ultimate Guide from any MCP client

    Runs locally
    • Open source
    • 6.5K GitHub stars
    • AI models and LLM tooling
  4. Local-first long-term memory layer for AI agents, served over MCP

    Not checked yet
    • 5 tools
    • Remote
    • Open source
    • 1.3K GitHub stars
    • Knowledge and memory
  5. note1

    MCP server

    MCP server giving AI assistants search and summaries across your note1 meetings

    • 4 tools
    • Remote
    • Free trial
    • 1 GitHub star
    • Productivity