These are instructions for you (the coding agent) to connect Dayze — Life Context — to this session. This is official guidance from Dayze; commands below are ready to use.
Dayze owns two agent lanes:
- Public temporal/biographical context: notable_pack / compare_lives (who is X, age in days, day_number milestones, birthday peers). Not an AI identity registry.
- Private Life Context Protocol: get_context_pack for a signed-in user's life graph.
There are a few actions to complete. Do all of them and continue if any one has issues.
- wire Dayze MCP into this agent
- identify whether the user needs public notable packs or private life-graph context
- for private context, help them create an API key or complete OAuth at Settings → Agent access
- optionally install the Dayze CLI for terminal harness use
IMPORTANT: Prefer doing setup steps yourself where the environment allows (edit MCP config, run CLI install). Only ask the user for browser login / API key paste when required.
---
## 1. Wire Dayze MCP (Streamable HTTP)
Add Dayze as an HTTP MCP server. OAuth metadata is auto-discovered from https://dayze.com.
Cursor (.cursor/mcp.json) or Claude Desktop:
{
"mcpServers": {
"dayze": {
"type": "http",
"url": "https://dayze.com/api/mcp"
}
}
}
Claude Code:
claude mcp add --transport http dayze https://dayze.com/api/mcp
Codex:
codex mcp add dayze --url https://dayze.com/api/mcp
ChatGPT Apps / Connectors: set connector URL to https://dayze.com/api/mcp
Gemini CLI / Google ADK (Python):
from google.adk.tools.mcp_tool.mcp_toolset import MCPToolset, StreamableHTTPConnectionParams
toolset = MCPToolset(
connection_params=StreamableHTTPConnectionParams(
url="https://dayze.com/api/mcp",
)
)
REST compat (optional): https://dayze.com/api/v1/mcp
MCP version: 1.9.0
## 2. Auth
Public tools (no login): notable_search, notable_profile, notable_pack, notable_pack_premium, compare_lives
Private life graph: Authorization: Bearer dayze_k_… (API key) or OAuth dayze_at_… (scope context)
Share tokens (read-only get_context_pack | get_life_graph): Bearer dayze_share_…
Create keys / share grants / manage connected apps: https://dayze.com/settings/agent-access
Scopes: mcp, context, offline_access — use actions scope for CLI write tools against the Agent stream.
## 3. First tool call (prefer one-call bootstrap)
Public who-is-X (no key) — prefer notable_pack over notable_profile:
curl -X POST https://dayze.com/api/mcp \
-H 'Content-Type: application/json' \
-H 'Accept: application/json, text/event-stream' \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"notable_pack","arguments":{"slug":"albert-einstein"}}}'
Compare two lives:
curl -X POST https://dayze.com/api/mcp \
-H 'Content-Type: application/json' \
-H 'Accept: application/json, text/event-stream' \
-d '{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"compare_lives","arguments":{"slug_a":"elon-musk","slug_b":"steve-jobs"}}}'
Private user life graph — prefer get_context_pack:
curl -X POST https://dayze.com/api/v1/mcp \
-H 'Authorization: Bearer dayze_k_YOUR_KEY' \
-H 'Content-Type: application/json' \
-d '{"tool":"get_context_pack","parameters":{"query":"this week"}}'
## 4. Optional — Dayze CLI (terminal harness)
Same agent tools as the web chat, from the Dayze repo:
cd dayze-webapp-main/cli && npm install && npm run build && npm link
dayze login --key 'dayze_k_…'
dayze # interactive session
dayze ask "…" # one-shot
CLI docs: https://dayze.com/docs/agents
API base override: DAYZE_API_BASE (default production). Actions scope required for write tools on the Agent stream.
## 5. Discovery & authenticity
Human connect UI: https://dayze.com/mcp
Full guide: https://dayze.com/AGENTS.md
LCP shape: https://dayze.com/docs/life-context
Docs: https://dayze.com/docs/agents
https://dayze.com/.well-known/mcp.json
https://dayze.com/.well-known/mcp/server-card.json
https://dayze.com/openapi.json
https://dayze.com/llms.txt
You only need to fetch https://dayze.com/ai during initial setup or when refreshing Dayze agent context. Prefer get_context_pack for session bootstrap of a signed-in Dayze user.
Dayze — Life Context
Dayze
Let your AI use Life Context
Give coding agents your days, people, and memories — via MCP — so they act with continuity, not chat history alone.
Start connecting
Copy this instruction and give it to your agent
fetch https://dayze.com/ai to connect Life Context
Or open to read the playbook yourself.
FAQ
Is it free?
Public notable-people tools have a free tier, then pay-per-call via x402 (USDC on Base). Private life-graph tools need a Dayze account + API key or OAuth — create one at Settings → Agent access.
Do I need a Dayze account?
Not for public packs (e.g. Einstein life-in-days). Yes for your private calendar, people, and memories — or use a read-only share token.
What agents does this work with?
Claude Code, Cursor, Codex, ChatGPT connectors, Gemini CLI / ADK — any agent that can fetch a URL and use MCP.
What does my agent get?
Life Context Protocol: get_context_pack for your graph, plus public notable_* knowledge packs. Optional Dayze CLI for the same in-app agent tools in a terminal.
How is this different from dayze.com chat?
Web / CLI Dayze Agent is the first-party harness (tools that write your life graph). This page wires other coding agents into Dayze via MCP — same idea as Netlify's agent bootstrap.
Life Context is structured, longitudinal data about a person’s days, people, events, money, and memories — so agents can act with continuity, not chat history alone. Connect UI · Docs · AGENTS.md