Moosewave MCP
Bring Moosewave into the AI tools you already use.
Model Context Protocol, or MCP, is a standard connection that lets a compatible AI assistant use approved tools from another product. Moosewave MCP makes scoped marketing work available without turning the assistant into a hidden workspace admin.
Ask your preferred assistant to inspect a campaign, prepare a draft, check deliverability, preview a migration, or review a result. Moosewave still checks the workspace, permission, input, and approval rules before the work continues.
You ask
“What should we fix before tomorrow’s campaign?”
- 01ReadCampaign context
- 02CheckInbox evidence
- 03ReturnRecorded answer
A connection, not another chatbot
Keep your preferred assistant. Give it a clearer way to work.
Normally, an assistant knows only what you paste into a chat. MCP gives it named, structured tools: read this campaign, prepare that draft, check this audience, or request that action. The result comes from the product rather than from guesswork.
Moosewave stays in the middle as the authority. It decides what the connection may see, validates every request, applies the rules of the workspace, and records what happened.
- 01
Ask in plain language
Begin with the outcome, not an API manual.
- 02
Review work in Moosewave
Drafts and evidence land where the team already works.
- 03
Follow every call
The tool, connection, timing, and outcome stay visible.
Useful work, not AI theatre
Start with a job your team already does.
Choose an example, then move through it one step at a time. The assistant asks; Moosewave checks; the workspace records; your team keeps the consequential decision.
Read-only investigation
Turn a vague result into a useful next step.
Your assistant can open the campaign, read its inbox-placement evidence, check the sending setup, and organise the findings around what your team can actually change.
This investigation can remain entirely read-only.
Why did our last campaign struggle?
Read the campaign, audience, and delivery record
Moosewave returns the selected campaign and its recorded result, so the answer starts from the right send rather than a pasted screenshot.
The assistant is looking at the requested campaign
- Open the campaignWorking now
- Check inbox evidenceWaiting
- Rank the next fixesWaiting
Three places to begin
The question is simple. The work behind it is real.
Every example begins with a normal request, names the steps Moosewave can perform, and states the boundary before the payoff.
- 01
Read-only investigation
Diagnose a campaign
“Why did our last campaign struggle?”
Your assistant can open the campaign, read its inbox-placement evidence, check the sending setup, and organise the findings around what your team can actually change.
- 01Open the campaign
- 02Check inbox evidence
- 03Rank the next fixes
BoundaryThis investigation can remain entirely read-only.
Why it helpsA diagnosis you can verify—not a generic copy critique.
- 02
Draft before impact
Prepare a journey
“Prepare a welcome journey in our voice and show me every message before anything is scheduled.”
The assistant can use your saved voice, templates, and audience rules to prepare a journey inside Moosewave, where the team can edit the actual drafts.
- 01Read the brief
- 02Prepare the drafts
- 03Review in Moosewave
BoundaryCreating a draft and sending it are separate actions. Campaign sends follow your team’s approval policy.
Why it helpsUseful work arrives in the product instead of ending in a chat transcript.
- 03
Discover before commit
Preview a migration
“Show me what would move from our current platform before we commit.”
The assistant can help inventory a supported source, narrow the audience, run a dry preview, and explain exclusions and verification evidence before a reviewed commit is requested.
- 01Discover the source
- 02Run the dry preview
- 03Review before commit
BoundaryDiscovery and dry preview do not write destination data. Commit is a separate, higher-permission action.
Why it helpsA migration your team can inspect instead of a blind import.
One connection, broad product reach
The tool catalog follows the work, not the org chart.
Moosewave MCP spans the connected marketing workflow. The tools available to one connection still depend on its scopes, write setting, workspace policy, and the product surface in use.
Campaigns & templates
Find and adapt templates, prepare drafts, run preflight checks, compare campaigns, and inspect the result.
Audiences & consent
Search subscriber context, preview segments, work with lists and tags, and keep suppression above targeting.
Deliverability
Check authentication, reputation, links, rendering, and inbox-placement evidence before deciding what to change.
Journeys & channels
Prepare automation flows, email, SMS, forms, and send-time decisions with the right channel rules in force.
Analytics & integrations
Read campaign results, compare evidence, inspect connection health, and bring outside events into the same workflow.
Migration
Discover, dry-run, verify, commit, monitor, and prepare rollback evidence for supported platform moves.
Go deeper into how Moosewave handles deliverability, analytics, automation, and audience control, or follow the connected ecosystem from idea to result.
Useful access, visible boundaries
MCP is not a shortcut around your workspace.
It is powerful because the assistant can request real product tools. Moosewave treats that connection like production access: bounded, checked at execution, and visible afterward.
- 1Access
A key with a small job
Use a dedicated workspace key for each connection. Start read-only, keep writes off by default, and grant only the scopes the job needs.
- 2Checks
The same product rules
MCP calls use Moosewave validation and workspace authorization. An AI client does not become a hidden administrator or a way around suppression.
- 3Decision
A deliberate point of impact
Drafting and sending remain different actions. Sandbox mode can simulate supported changes; sensitive work can wait for review; sends follow team policy.
- 4History
A trail back to the request
Each call records its tool, connection identity, outcome, and timing. Important changes also remain visible through the product’s audit history.
From question to recorded result
Five steps. No mystery handoff.
Connection details and confirmed client compatibility are provided during private early-access onboarding, so the setup matches the runtime your team actually uses.
- 01
Choose a connection.
Use a compatible desktop assistant, editor, remote client, or a private team runtime.
- 02
Choose what it may do.
An administrator gives the connection a dedicated workspace key with only the access it needs. Write access can stay off.
- 03
Ask for an outcome.
The client selects a typed Moosewave tool. Moosewave validates the input and checks workspace permission before it runs.
- 04
Review before impact.
Read-only work can return immediately. Supported writes can be simulated, and consequential work follows the approval rules in force.
- 05
Follow the history.
The request, tool, key, timing, and outcome stay visible. Revoke the connection whenever it no longer needs access.
Local connection
For compatible desktop assistants and editors. The connection runs alongside the client.
Remote connection
For compatible clients that can connect to an authenticated MCP service.
Private team runtime
For shared agents and internal tools running inside your own authenticated environment.
The AI-provider boundary
Moosewave controls workspace access. Your chosen AI client controls the model side.
Moosewave decides which tools a connection may request and what authorized results it receives. The client or model provider you choose decides how prompts and returned information are processed, retained, or used. Review those privacy and retention settings as carefully as the workspace key.
Review Moosewave’s current security and data-protection posture before connecting a production workspace.
Questions teams ask before connecting MCP
The short version: start with a bounded read-only job, use a dedicated key, and make the path from request to result easy to inspect.
What is Model Context Protocol (MCP)?
MCP is a standard way for an AI client to use tools from another product. Moosewave MCP makes permitted campaign, audience, deliverability, automation, analytics, integration, and migration actions available to compatible clients.
Do I need to be a developer to use Moosewave MCP?
Not for everyday use. An administrator or developer makes the initial connection; after that, a marketer can ask for work in plain language and review the result in Moosewave. Technical teams can also manage local or private connections themselves.
Which AI clients can connect to Moosewave MCP?
Moosewave supports compatible desktop, editor, and remote MCP clients, plus private HTTP connections for team runtimes. Current client-specific setup is confirmed during onboarding because support also depends on the MCP features offered by the client you choose.
Can an MCP client send a campaign or change customer data?
Only when its workspace key has the required access and Moosewave allows the action. You can keep writes off, begin with read-only access, simulate supported changes in sandbox mode, and keep campaign sends behind your team’s approval policy.
Does MCP bypass Moosewave approvals?
No. When a consequential action is covered by an approval rule, the request follows that approval path rather than skipping it. Not every write requires approval, so administrators should choose scopes, write access, and workspace policies deliberately.
Where does my data go when I use an AI client?
Moosewave returns only the information permitted for that connection and records the tool call. The AI client you choose controls how prompts and returned information are processed, retained, or used, so its privacy and retention settings matter too.
Can our team run the MCP connection privately?
Yes. The server supports local connections and an HTTP transport that can run inside a private, authenticated environment. Connection details are provided during Moosewave onboarding while the product remains in private early access.
Does MCP replace the Moosewave API or other developer tools?
No. MCP is another way to request Moosewave actions through a compatible AI client. Technical teams can still choose the API, SDK, CLI, webhooks, or MCP for each workflow.
Give your assistant something useful to do.
Explore the interactive product, then join the waitlist. During onboarding, we will help you choose a compatible connection and begin with the smallest useful permission set.