Roleplane— the context platform for your team's AI

In development · piloting with first teams

A smart model with no context is just an expensive guess.

Roleplane is the context management platform for your team's AI. Connect your docs, channels, repos and servers once, build the context each role needs for its job, and deliver it to Claude, ChatGPT and Cursor through one link.

Every assistant on your team is guessing at your business.

Build the context once, shape it per role, and deliver it to any assistant through one link.

An on-call AI is only as good as the logs it can see.

Logs, status and Kubernetes state become part of the context each role works with. Read-only, a separate user per role, every command logged, no SSH keys on laptops.

Every role gets its own plane

Drop in Notion, Atlassian, Slack and the rest. Each role's context is assembled from the sources its job needs, then sent out to its assistant.

What breaks when AI works without your context

01

The AI starts every task blind

A fresh chat knows nothing about your product, your decisions or last week's incident. People paste the same background again and again, and the model fills the gaps with a confident guess.

02

One pile for every role

Connect Notion or Slack to Claude or ChatGPT and every role gets the same pile: the engineer's AI wades through sales decks, support's through code. The model loses the thread, and what the job needs is buried. Teams describe the choice as connecting everything or nothing.

“It's all or nothing.”A Hacker News comment

03

Each assistant has its own picture

Every assistant gets its own hand-pasted description of the company, so the pictures drift apart and nobody knows which one is current.

04

Every person repeats the setup

Each person authorizes every tool and keeps keys on their own laptop: OAuth, API keys, kubeconfigs, SSH keys. A new hire starts from zero.

“MCP auth has been a huge pain point for us.”Another Hacker News comment

From scattered sources to role-ready context

Roleplane sits between your tools and your team's assistants and does four jobs.

01

Connect

Read-only links to Notion, Slack, Google Drive, Trello, GitHub, Linear, servers and Kubernetes. Others on request.

  • Notion
  • Slack
  • Drive
  • Trello
  • GitHub
  • Linear
  • Servers
  • Kubernetes

02

Shape

Pick pages, channels, repos and hosts. Trim, redact, summarize and restructure them into context sized for a role.

  • Select
  • Trim
  • Redact
  • Summarize
  • Restructure

03

Assemble

Bundle the context with the role's rules and skills, so the AI knows what to read and how to behave.

  • Context
  • Rules
  • Skills

04

Deliver

One link per role for Claude, ChatGPT, Cursor and any MCP client. Live sources stay current; exported packs get a new version.

  • Claude
  • ChatGPT
  • Cursor
  • MCP

Same company, different context per role

Each role's AI works from its own branch of your company context: what its job needs, without the rest.

Animated example: three roles each get their own part of one company context tree, and what is closed stays closed.

What else the platform does

Beyond choosing sources, a role's context needs shaping, rules and a way to check the result.

Shape the content for the job

Trim long pages, summarize threads and redact salaries so the AI gets what the task needs, not a pile.

Rules travel with the context

Each role gets instructions: what to check first, what never to do. They go with the connector or as a skill.

One link, every assistant

Claude, ChatGPT, Cursor and any MCP client load the same role context. New hires get it on day one.

A log of what the AI did

Every command on your servers is logged, and before handover you get the log of refused actions. A log for documents is in development.

See as this role

Preview the context a role's AI will work from before you hand it over.

Change it in words

Say what to change, review the diff, apply. Nothing changes until you confirm.

How others handle your team's AI context

From public docs, September 2026. Each has real strengths; this is where teams usually run into a limit.

ApproachWhat it does wellWhere it stops
Claude and ChatGPT admin controlsVendor controlsIncluded in the seat price. Admins allow, approve or block connectors. Claude can also set each tool of a connector separately, and ChatGPT can limit an app to read actions. For Google Drive on Enterprise and Edu, ChatGPT can also restrict the synced source to chosen shared drives or folders.Control stops at the connector and its actions, inside one vendor. The docs describe no way to give a role only chosen Notion pages or Slack channels, to redact, trim or restructure what comes back, or to build one pipeline across sources. In Claude, custom roles are Enterprise-only.support.claude.com ↗ help.openai.com ↗
MCP gateways: Cloudflare Portals, Obot, WillowMCP gatewaysOne login, audit logs, large connector catalogs. Cloudflare Portals are free up to 50 seats; Willow is free up to 5 users.They control which servers and tools a role can call, not which pages those tools return. Willow's content filter masks personal data.blog.cloudflare.com ↗ obot.ai ↗ withwillow.ai ↗
DDustAgent platformAn agent platform. Admins pick the exact pages and folders per Space, and since June 2026 Claude and Cursor can search them over MCP.The MCP server searches the Spaces a user can open; to use an agent's instructions you talk to that agent inside Dust. Free, Pro and Max plans have up to 5 Spaces, and there is no built-in connection to Trello, Linear, servers or Kubernetes.docs.dust.tt ↗ docs.dust.tt ↗
CCredalGoverned MCPBundles curated data, tools and instructions into a governed MCP server per department.Sold through an enterprise sales process with no public price, built for platform and IT teams.credal.ai ↗
GGleanEnterprise searchPermission-aware search across 100+ sources, with an MCP server for every assistant.An enterprise deployment; MCP pricing is not public.docs.glean.com ↗

Where Roleplane fits

Teams of 10–50

  • Build each role's context from exact pages, channels, repos and hosts, then trim, redact or summarize it, not only allow a connector.
  • Role rules travel with the context.
  • Servers and Kubernetes, read-only, with a command log.
  • Set up by hand in a week for $150/month per team; you pay after it works.

Larger teams

  • One role setup for Claude, ChatGPT, Cursor and any MCP client, instead of separate admin settings in each.
  • Add a source or a role once; everyone in the role gets it through the same link.
  • No OAuth apps, keys or kubeconfigs on individual laptops; new hires add one link.
  • Preview the context a role's AI will get and review changes as a diff before they apply.

Pilots are teams of 10–50 today. Larger team? Tell us your size.

Spot a mistake? Write to [email protected] and I'll fix it within two days.

How it works

  1. Connect your sources, read-only: Notion, Slack, Google Drive, Trello, GitHub, Linear, servers, Kubernetes.
  2. Shape the context for each role and add its rules: which pages, channels, repos and hosts, trimmed, redacted or summarized as needed.
  3. Each person adds one link to their assistant. New hires get their role's context on day one.

Piloting now

I set up Roleplane for your team by hand, within a week: your sources, a context for each role, one link per assistant.

  • Live, read-only access to what you pick in Notion, Slack, Google Drive, Trello, GitHub and Linear.
  • Servers and Kubernetes, read-only: a separate user or service account per role, a whitelist of commands, no access to secrets, every command logged, keys kept off laptops.
  • On request, with your admin: Jira and Confluence (paid Atlassian plans), SharePoint, HubSpot, Zendesk and read-only database views.
  • Anything else — legal docs, runbooks, wiki exports — as a role's knowledge pack, updated by hand.
  • Access is created on your side and deleted when the pilot ends; we agree on this in writing.
  • Works with Claude (custom connectors, including Team and Enterprise), Claude Code, Cursor and any MCP client. ChatGPT connects through custom connectors: developer mode on Plus and Pro, enabled by an admin on Business and Enterprise. Read-only is all it needs to load context.

Founding price: $150/month per team of up to 25 people, locked for 12 months.

You pay only after it works. Larger team? Tell us your size.

Request a pilot

Questions

Why does each role need its own context?
The same model gives a useful answer or an expensive guess depending on what it knows about your work. A shared pile buries what a role needs, and nothing at all leaves it guessing. Roleplane builds the context for a role: the pages, channels, repos and servers its job needs, shaped to fit, with the role's rules on top.
Don't Claude and ChatGPT already have roles?
They control which connectors and tools a role can use, inside one vendor. In Claude, custom roles are Enterprise-only. Roleplane builds what each role's AI works from inside those sources — the pages, channels, folders and hosts for its job, shaped and with its rules — and delivers it to any assistant.
Is this a gateway or a search tool?
Neither. A gateway decides which tools an assistant may call, and search finds documents. Roleplane builds the context a role works from: which sources, shaped how, with which rules, delivered to any assistant.
Which assistants does it work with?
Claude, ChatGPT and Cursor load the role context directly. Gemini, DeepSeek and other models work through MCP clients such as Gemini CLI and Cline.
Isn't this what Credal does?
Yes, Credal scopes data per role too — for large companies, through a sales process. Roleplane is for teams of 10–50: set up by hand within a week, $150/month per team.
We already use Dust.
Dust's MCP server lets Claude and other clients search the Spaces a Dust user can access. If your team mostly needs company knowledge inside Claude, ChatGPT or Cursor rather than Dust's own agents, Roleplane gives each role its pages for $150/month per team, including sources Dust has no built-in connection for: Trello, servers and Kubernetes. If you build agents in Dust, keep Dust.
We already have an MCP gateway.
Gateways decide which tools a role can call. Roleplane decides which pages, channels and folders those tools return, as one more MCP server per role next to your gateway.
Can't we set up one integration or bot per role ourselves?
Yes — that's how the pilot works under the hood. What you get: someone sets it up and keeps it in sync, one place to see what each role's AI can reach, and the same setup for every assistant.
Which tools can the pilot connect?
Live and read-only: Notion, Slack, Google Drive, Trello, GitHub, Linear, servers over SSH and Kubernetes. On request, with your admin: Jira, Confluence, SharePoint, HubSpot, Salesforce, Zendesk, Intercom, GitLab and read-only database views. Anything else goes in as a role's knowledge pack. Before handing over access, we try what each role's AI must not do and send you the log of refusals.
Is it safe to let an AI read our servers?
Pilots are read-only. You create a separate user or Kubernetes service account for each role; it can run only the whitelisted commands (logs, status, the configs you pick) and cannot read secrets. The key lives on our server, not on laptops, and every command is logged. Delete the user and access is gone.
Is it safe to give you access to our docs?
Read-only, only what you pick, access deleted after the pilot, agreed in writing before we start.
What happens after the pilot?
If it works, you keep it at the founding price. If it doesn't, you pay nothing and we delete the access.

Request a pilot

Which AI assistants does your team use?

Giving each role's AI its own context (specific pages, channels, repos, servers) is…

Where does the context your team's AI needs live? (optional)

We store your email and answers only to reply and set up a pilot, and delete them on request. No cookies. Privacy policy.

Not ready for a pilot? Get one email when Roleplane opens.