Post to social media from Claude

Custom connectors are on every Claude plan, with one limit on Free

Anthropic’s help center says custom connectors using remote MCP are available “for users on Free, Pro, Max, Team, and Enterprise plans”, and that “Free users are limited to one custom connector”.[1]

By that rule a Free account can hold the Postnavo connector as long as it is the only custom one.

To change a connector you remove it and add it again with the new details. The address on this page is illustrative, so add the connector once the final one is published at launch.[1]

On Team and Enterprise an owner adds it and members connect

Only Owners and Primary Owners can add a custom connector to a Team plan. On Enterprise, people with a custom role that includes “Manage access to Libraries” can add one too.[1]

The owner works in Organization settings, Connectors: “Add”, hover over “Custom”, select “Web”, then the name and the server URL.[1]

After that each member goes to Customize, Connectors, finds the connector with its “Custom” label and clicks “Connect”. Anthropic’s reason is that Claude should only reach the tools and data the individual user can reach.[1]

Postnavo is launching with OAuth sign-in, so in an agency each teammate connects their own Postnavo account.

Claude reaches the server from Anthropic’s cloud

When you add a custom connector, Claude connects to the server “from Anthropic’s cloud infrastructure, rather than from your local device”. Anthropic says that is true on every client, including claude.ai, Claude Desktop, Cowork and the mobile apps.[1]

A server on a private network or behind a VPN will not connect. Postnavo is launching with a hosted MCP server, so there is nothing to run or open up on your side.[1]

Connectors you add here also show up in Claude Code when you log in there with the same claude.ai account. The Claude Code page covers that side.[4]

What Claude asks before it posts

A connector’s tools can read data, create or change it, and act for you. Anthropic’s advice is to review Claude’s tool approval requests carefully and to click “Allow always” only for a server and tool you trust to run unsupervised.[1]

You can switch off tools you do not want in a conversation from the “Search and tools” menu.[1]

Research is the exception. During research Claude can call connector tools “automatically without further approval”, and Anthropic tells you to disable tools that take write actions in external applications.[1]

Postnavo is launching with an approval setting of its own. With it on, a post Claude creates waits in Postnavo for your OK even after “Allow always”.

Signing in from Claude, and taking access back

Adding a connector usually runs an OAuth sign-in, which lets Claude act in the application for you “without Claude ever seeing your actual password”.[1]

Anthropic advises reviewing the permissions a server asks for during that sign-in, and denying access if they seem unnecessary. You can revoke them at any time by disconnecting the connector in Claude’s settings, or in the other service’s security settings.[1]

Postnavo is launching with OAuth sign-in for this reason: access is granted per person and can be withdrawn.

Scheduled tasks and skills in Claude

Claude can run tasks on a schedule on all paid plans: Pro, Max, Team and Enterprise. Scheduled tasks run remotely, so they keep their cadence while your computer is asleep, and they can use your connectors. The frequencies are hourly, daily, weekly, on weekdays, or manual.[3]

A weekly task could draft next week’s posts and hand them to Postnavo. The posts do not need a Claude task to go out: once scheduled, each one sits in your Postnavo calendar, which is launching with the app.

Skills are a separate feature, on Free, Pro, Max, Team and Enterprise with code execution enabled. You add one under Customize, Skills by uploading a ZIP of the skill folder.[2]

Postnavo’s agent skill is launching for its CLI and for coding agents. In Claude’s chat apps the connector is the route.