Who it is for
Teams keeping one source of truth that both people and agents read and write.
Coconut Context
A versioned markdown context layer shared by people and agents
Overview
What Coconut Context does
Coconut Context addresses the cold-start problem: every new chat and every new agent begins without knowing the customers, decisions and house style already established, so someone re-explains them.
The fix is a versioned knowledge base written in plain markdown, readable and writable by both people and agents. Canonical context covering product docs, customer accounts, processes and decisions can be searched and read instead of guessed at, and written back. The answered surface is twice the claimed size, adding spaces for organisation, queryable page metadata, and revision history that makes changes traceable. Community tier.
Tool surface
Thirty tools answered against fifteen claimed. Beyond search and read sit spaces such as context_space_list and context_create_space, page metadata with queryable keys, revision history, and context_whoami for identity.
Availability
Where Coconut Context is listed
Coconut Context is listed in the Claude connectors directory only. Coconut AI has no app under the same domain in the ChatGPT directory; 392 of the 3,150 products here are on both marketplaces, and this is not one of them.
The MCP server behind a Claude connector is usually the same server a ChatGPT app would need, so the absence is a distribution decision rather than a technical one.
Claude listing
In the Claude connectors directory
Coconut Context publishes 15 tools to Claude. That is a mid-sized surface: 59.4% of the 1,253 Claude-listed connectors expose fewer.
The median Claude-listed connector filed under Productivity publishes 12, so this one runs 3 tools above its bucket's median and ranks 199th of 462 by tool count.
It falls in the 11-20 tools band, which holds 22.5% of the Claude directory.
Measured against the live endpoint
A direct call to the endpoint returned 30 tools, 15 more than the 15 the listing publishes. Probed 2026-08-07.
Tools the server actually returns
context_whoamicontext_space_listcontext_create_spacecontext_create_shared_pagecontext_get_shared_pagecontext_get_page_metadatacontext_get_page_linkscontext_set_page_metadatacontext_query_pages_by_metadatacontext_list_metadata_keysget_context_by_urlcontext_list_shared_pagescontext_recent_shared_pagescontext_get_shared_page_revisionscontext_search_shared_pagescontext_update_shared_pagecontext_create_personal_pagecontext_list_personal_pagescontext_get_personal_pagecontext_update_personal_pageagent_listagent_list_tasksagent_get_taskagent_create_taskagent_update_taskagent_run_taskagent_list_runsagent_get_runagent_get_instructionsagent_set_instructions
Coconut Context publishes a remote MCP endpoint. Claude connects to it over HTTP; there is nothing to install locally.
Every connector in the Claude directory is remote, all 1,253 of them, so transport is not a differentiator there.
- Publisher
- Coconut AI
- Verification
- Community-listed (61.5% of the Claude directory)
- Filed under
- Productivity
- Endpoint
-
https://api.coconut.md/mcp - Documentation
- Publisher setup docs
- Published tools
- 15 · 11-20 tools band
- Depth rank in Productivity
- 199 of 462
- Snapshot
- August 2026
Coconut Context is community-listed in the Claude directory, the tier 770 of 1,253 connectors sit in (61.5%). It is the default for a self-published server, and says nothing either way about quality.
In the publisher's words
How Coconut AI describes it
Coconut Context is the shared, living context layer for your organization, built so humans and AI agents work from the same source of truth. Most AI tools forget. Every new chat, every new agent starts cold, and you re-explain your customers, your decisions, your product, your house style. Coconut Context fixes that. It's a versioned knowledge base written in plain markdown, readable and writable by both people and agents, and exposed to Claude over MCP. Connect it once and Claude can: - Search and read your canonical context (product docs, customer accounts, processes, decisions) instead of guessing - Write back what it learns, so knowledge compounds across sessions instead of evaporating - Work in shared spaces (team knowledge) or personal spaces (your own working memory), with permissions that keep the right context in the right hands - Stay current through versioning, so every change is tracked and reversible Because it's plain markdown, your context stays portable and human-readable, not locked in a proprietary format you can't inspect. Humans edit it in the Coconut Context app; agents read and update it through the connector. This is how you ship at agent speed without the slop: give your AI durable, permissioned, organizational memory instead of re-priming it every time. Coconut Context is part of the Coconut platform for building and running autonomous software agents. Use it on its own, or alongside Coconut's agent runtime and dashboard.
Quoted from the connector's own Claude directory listing as captured in August 2026. Node8 did not write it and does not vouch for it.
Nearby
Other Productivity connectors
Basic Memory Cloud
Persistent Markdown knowledge base shared across AI tools
Claude · ChatGPT
UseMyContext.ai
Read-only personal context with cited document answers
Claude · ChatGPT
AgentContext
Business context from a data stack: schemas, concepts and metadata
Claude
Great Arrow Digital
A private workspace memory shared across several AI assistants
Claude
skilder
Versioned skills and MCP tools distributed from one governed hub
Claude
Upfluence
Creator discovery and campaign tracking in chat
Claude · ChatGPT
This is an independent catalogue entry compiled by Node8 from Coconut Context's public marketplace listing as it stood in August 2026. Listing facts come from the marketplaces; the comparisons against the rest of the directory and any live-endpoint measurements are Node8's. Node8 is not affiliated with Coconut AI, and inclusion here is not an endorsement.
Want your company reachable inside ChatGPT and Claude?
Node8 builds MCP servers end to end: deciding which tools are worth exposing, the authentication and permission model, the review submissions, and the onboarding that gets them used. One server serves ChatGPT, Claude, and Microsoft Copilot.