Browse documentation ⌄
MODEL CONTEXT PROTOCOL
Give your AI agent real social tools.
Connect an MCP-compatible client to Postzivo’s hosted endpoint and let it work inside the same roles, scopes, and approvals as your team.
Hosted endpoint
Streamable HTTPCopy and customize
https://app.postzivo.com/mcpClient configuration
JSONCopy and customize
{
"mcpServers": {
"postzivo": {
"url": "https://app.postzivo.com/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}i
Use a dedicated key per agent.
This isolates scopes, rate limits, rotation, and audit activity.
Recommended permissions
Content assistant
Draft, read brands, list media, and prepare campaigns without publishing permission.
posts:read · posts:writePublishing agent
Create or schedule posts with approval rules enabled for the workspace.
posts:write · accounts:readAnalytics agent
Read posts, performance, trends, creators, and competitors.
posts:read · analytics:readInbox agent
Read conversations and draft replies; keep response permission separate where possible.
inbox:read · inbox:writeOperational guidance
- MCP calls share the underlying API key’s plan rate limits.
- Require approval before publishing when an agent can generate autonomous content.
- Give each integration only the scopes it needs.
- Rotate a key immediately if it appears in chat history, logs, or source code.
READY TO BUILD?
Connect your workflow.
Request access, create an API key, and start with one test workspace.