AI & Machine Learning

ctxstore.ai

An AI agent memory server that stores and retrieves facts, binds agent identities to sessions, and manages namespace invitations.

1 endpoint31 known toolsFirst detected October 6, 2026Last detected October 10, 2026

ENDPOINT 1

https://mcp.ctxstore.ai/mcp

No auth detected

MCP server metadata

Name: emily-context
Version: 1.0.0

Capabilities

experimentalresourcestools.listChanged

Server instructions

This account has persistent memory on ctxstore. What is stored here is available next session, in any client, to any agent the account holder connects. The memory arrives on its own: the first tool result of every session opens with it, addressed as the memory of the person you are assisting. Nothing needs fetching first. How people get value from it, in the order it usually happens: 1. If a tool answers with a setup prompt, the account is not linked yet. setup_account(mode='link_or_create', email) sends a six-digit code; setup_account(mode='verify', code, email) finishes. Keys are never pasted, echoed, or stored. 2. store_fact when a decision, a learning, or a change of state happens, in the turn it happens. Short, searchable, with a stable key when the fact will change later. 3. search_facts or get_fact before asking the person to repeat something they have told an agent before. 4. Near the end of a session, store_session_summary, then agent:<id>:self-note:closing, so the next session starts where this one stopped. 5. Bring their memory with you: if you already remember things about this person from your own platform memory, offer to carry them over with store_facts, one fact per item, in their words, with the date learned when known. An agent identity is assigned when the session starts (the memory block names it); bind_agent(agent_id=...) renames it when the person prefers a name. Full manual: https://ctxstore.ai/llms.txt

Known tools 31

accept_invite

Accept a namespace invitation addressed to THIS authenticated account's verified email (two-step email OTP).

Potential side effects

bind_agent

Bind a sticky agent identity to this MCP session.

Inferred read-only

decline_invite

Decline a pending (status=invited) namespace invitation as the grantee.

Inferred read-only

delete_fact

Soft-delete a stored fact by its UUID (Smart Forget).

Potential side effects

get_fact

Retrieve a fact by its exact key — deterministic, never misses.

Inferred read-only

get_fact_history

Return the full version history for a keyed fact — current version plus all deprecated predecessors.

Inferred read-only

get_session_seed

Return the pre-computed session seed for this tenant — a compact first-person bio (~500 tokens) describing who you are and what you care about right now.

Inferred read-only

get_stats

Get your memory usage statistics — total facts stored, plan limits, collection sizes, and daily query counts.

Inferred read-only

ghost_mode

Enable or disable ghost mode for this session.

Inferred read-only

index_recent

Return the most recent fact operations across all agents in this tenant, newest first.

Inferred read-only

index_search

Filtered search over the tenant activity index.

Inferred read-only

index_stats

Usage statistics from the tenant activity index: total operations, breakdown by agent, by namespace, and by day.

Inferred read-only

introduce_yourself

Optional: introduce yourself to the platform.

Inferred read-only

leave_grant

Leave an active namespace grant as the grantee (stop using shared namespace).

Inferred read-only

list_facts_by_key_prefix

Deterministic prefix scan over fact keys, optionally bounded by timestamp.

Inferred read-only

list_invites

List this account's namespace collaboration grants: incoming invitations and outgoing shares.

Inferred read-only

list_workspaces

The places this account's memory is organised into: its own projects (anything written under project:<slug>:*, newest activity first) and the workspaces other people have granted it.

Inferred read-only

load_context

Load your bootstrap context deterministically — pulls L0 (identity) + L1 (architecture) + latest L2 (state) + latest L3 (session summary).

Inferred read-only

mark_salient

Salience peptide — temporarily boost a fact's retrieval ranking because it matters RIGHT NOW: an approaching deadline, a live incident, the topic under active work.

Inferred read-only

move_fact

Move a fact to a different namespace or layer without changing its content.

Inferred read-only

reinforce_fact

Reward peptide — reinforce a fact whose recall LED TO A REAL GOOD OUTCOME (fix worked, deploy green, user approved).

Inferred read-only

resume_session

Call to load session continuity.

Inferred read-only

revoke_grant

Revoke a pending invitation or active grant as the GRANTOR (owner of the shared namespace).

Inferred read-only

search_context

Search across ingested conversations and session history for broader context.

Inferred read-only

search_facts

Search your persistent memory for stored facts.

Inferred read-only

setup_account

Set up or restore your persistent memory connection.

Inferred read-only

store_fact

Store information that should persist across sessions.

Inferred read-only

store_facts

Store up to 50 facts in one call — the same fields as store_fact per item.

Inferred read-only

store_session_summary

Store a compressed summary of this session to preserve continuity for the next one.

Inferred read-only

switch_workspace

Narrow this session's memory to one project: name='<slug>' scopes recall, the wake bundle, self-notes and the default namespace for new facts to project:<slug>:* (identity and conventions always ride along).

Inferred read-only

wake_status

Check which context layers loaded successfully on this session start.

Inferred read-only

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.emily-context]
url = "https://mcp.ctxstore.ai/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "emily-context": {
      "type": "http",
      "url": "https://mcp.ctxstore.ai/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: emily-context
Remote MCP URL: https://mcp.ctxstore.ai/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "emily-context": {
      "url": "https://mcp.ctxstore.ai/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "emily-context": {
      "type": "http",
      "url": "https://mcp.ctxstore.ai/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "emily-context",
  "transport": "streamable-http",
  "url": "https://mcp.ctxstore.ai/mcp"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

TRUST AND VERIFICATION EVIDENCE

Loading Trust v2 evidence…

Checking the associated registrable domain. The BuiltWith key remains server-side.

Indexed

Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.