Introduction

Coco (Coconut Context) — a solution platform built on pages

Coco is a solution platform built on pages. A page is a markdown document that reads identically to a human in a browser and to an agent over HTTP or MCP, carries structured metadata (queryable key-value data), and lives in a space with real access control. Every space has one agent that can run tasks on demand or on a schedule, using connectors to outside systems.

That combination — narrative + structured data + an agent + integrations — is what turns a wiki into an application platform:

• The page body is the human-readable narrative (an investment memo, a runbook, an account brief).
• The metadata is the machine-readable state that churns (a stage, a score, a growing list of sources) — individually patchable, audited per key, and queryable across pages without touching the document.
• The space agent keeps both fresh: it folds new information into the prose and updates the metadata, on demand or on a schedule.
• Webhooks, export/import, and templates connect the result to the rest of your stack and make solutions repeatable.

The three doors

One page, one resolver, three doors — all governed by the same ACL engine:

Door	Who	How
Browser	Humans	Rendered HTML at /pages/:space/:path
HTTP API	Agents & systems	Accept: text/markdown (or .md) returns clean markdown; JSON endpoints for everything else
MCP	AI agents	context_* tools over JSON-RPC with OAuth or bearer auth

Pages are versioned in Postgres as append-only revisions — no git lifecycle. Every write is a new revision; historical reads are one query parameter away.

Explore the docs

Quickstart

Run the stack, create a page, attach metadata, query across pages.

Concepts

Organizations, spaces, pages, metadata, agents, templates, and more.

HTTP API

The routes agents and systems use to read and write.

Solutions

Worked examples: PE pipelines, SaaS accounts, runbooks, contracts, suppliers.