Maeve Social

› Which drafts are still waiting on media this week? ⚙ list_content { status: "draft", workspace: "lily-dia" } ✓ 6 drafts · 4 Instagram · 2 TikTok · 2 missing media › Publish the ring-stack reel now. ⚙ publish_now requires mcp:dangerous + confirmation → confirm with "publish_now:cnt_8f3k2" to proceed

MCP earns its place when an agent needs to know what is actually going on before it can help. It can list the drafts waiting for review, compare what is scheduled across platforms, see which media there is to work with, or pull the last thirty days of analytics, straight from the workspace over typed tools.

Reading the workspace needs mcp:read and workspace access. Creating drafts, scheduling, and media uploads need mcp:write and an editor or admin role, and nothing skips the validation the app already runs. The tools that go public or cannot be undone, publishing now, deleting, bulk media moves, sit behind mcp:dangerous and only run when the call includes the exact confirmation text for that action and target, like publish_now with the content ID.

A hosted Model Context Protocol server at the Maeve API. Connected clients get typed tools for real workspace tasks, with the risky ones kept on a short leash.

48 Typed Tools28 for reading, 14 for writing, and 6 high-risk ones. Which of them an agent actually sees depends on the scopes and policies approved for that connection.

What Agents Can ReadWorkspaces and integrations, platform capabilities, drafts and scheduled posts, content counts, campaigns, taxonomy, hashtag groups, media with folders, labels and usage history, grid tiles, analytics, demographics, and integration health.

What Agents Can WriteWith mcp:write, an editor or admin can create a draft, schedule existing content, update planning notes, run the two-step media upload, and tidy media folders and labels. None of it skips the validation the app already does.

Dangerous Actionsmcp:dangerous unlocks the tools that go public or cannot be taken back: publish_now, retry_content, archive_content, restore_content, delete_content, and the preview-first organize_media_assets.

Confirmation SafeguardsEvery dangerous action needs an exact confirmation tied to the action and its target ID. Server-side auto-confirm exists, but only for named actions a policy has deliberately switched on.

AuthenticationApprove the client in your browser: the authorization page checks PKCE, shows exactly which scopes are requested, lets you pick the organization, and only then hands over a scoped token. An API key still works for headless setups.

Org And Workspace ChecksA browser-approved grant is tied to one organization, but every call still names a workspace, and the server checks you can reach it before anything is read or changed.

Token Expiry And RevocationAccess tokens last six hours and refresh tokens rotate for up to ninety days. Approved clients show up under MCP Connections, where you can cut off access straight away.

Server-Side LoggingEvery tool call is logged: the tool, who ran it, how they authenticated, the client, scopes, organization, workspace, targets, confirmation mode, duration, and outcome.

Rate And Origin ControlsThe MCP endpoint is capped at 60 requests a minute and held to the same allowed-origin rules as the API. Downstream actions still answer to each platform's own limits.

Outside MCP On PurposeThe Inbox, approvals, and client review are separate surfaces. MCP also does not generate PDF reports, reorder grid tiles, or manage API keys. Those live in the app, REST API, or CLI.

MCP is not a separate automation island. It reads from and writes into the same product surfaces your team already uses, so agent help stays visible in the normal workflow.

Maeve MCP is a controlled AI-agent interface focused on the posting workflow for your Maeve workspace. It lets MCP-compatible tools inspect social media drafts, schedules, integrations, media, analytics, tags, hashtags, campaigns, and grid-planner items, and compose, schedule, and publish content, with write actions gated by scopes, roles, and confirmations.

An MCP server exposes product actions as typed tools that an AI client can call after authorization. In Maeve, those tools sit on top of the real publishing, media, and analytics systems instead of giving an agent a separate sandbox. MCP is the curated posting slice of the product; other surfaces stay in the app, REST API, and CLI.

Maeve provides setup copy for Claude Code, Codex, and Cursor or other JSON-configured MCP clients. Browser-approved MCP has been validated in the product guidance with Codex and Claude Code, and any compliant HTTP MCP client can connect if it supports the required authentication flow.

With mcp:read, an agent can list workspaces, integrations, integration capabilities and options, content, content counts, campaigns, taxonomy, hashtag groups, media, media folders and labels, media usage history, grid items, analytics summaries, aggregate metrics, post analytics, demographics, and integration health.

Yes, but only through the high-risk publish_now tool. The connection must include mcp:dangerous, the user must have editor or admin access, and the call must include exact confirmation text such as publish_now:<contentId> unless a server-side dangerous auto-confirm policy has explicitly allowed that action.

Yes. Dangerous tools are not exposed unless the connection has mcp:dangerous, workspace roles still apply, and each high-risk call requires exact confirmation text tied to the action and target ID. Tool allowlists and denylists can also hide specific tools globally or for an organization.

High-risk tools require a confirmationText value that exactly matches the server's expected action and target, such as publish_now:<contentId>, delete_content:<contentId>, or organize_media_assets:<operationKey>:<count>. If the text is missing or wrong, the tool returns a confirmation-required error with the expected text.

Yes. Every workspace-scoped tool checks access before it reads or writes. Read tools require workspace access; write and dangerous tools require editor or admin roles. Viewer access can inspect allowed data but cannot create, schedule, publish, or delete through MCP.

Does MCP Work With Analytics And Reports?

MCP can read analytics summaries, aggregate metrics, post analytics, audience demographics, and integration health. It does not generate PDF analytics reports in MCP v1; use the app, API, or CLI for report generation where that workflow is available.

MCP is focused on posting. It does not handle the Inbox, approvals, or client review; it does not generate PDF reports, run destructive or broad media operations beyond the preview-first organize tool, reorder or promote grid items, manage API keys, or bypass platform/API constraints. Those workflows live in the app, REST API, or CLI. Publishing still depends on the connected platform's API.

Free MCP that drives an audit of your marketing. Your AI connects, adsOS digs through your ads, email and site, and hands back a growth plan you can run today.

Publish, schedule, moderate your social inbox, and pull analytics across every major network directly from any AI assistant, via Eclincher.

Connect AI agents to project data, analytics, research, content planning, media generation, and growth workflows.

Connect AI tools like Claude, Cursor, and Codex directly to your CitedSpy workspace with the Model Context Protocol (MCP). Once connected, your AI assistant can read your brands, prompts, runs, citations, and analytics - and create prompts or trigger runs - without you copying data back and forth. You can ask for your AI visibility dashboard in plain language, find your weakest prompts and engines, create and edit tracking prompts from chat, trigger fresh runs across engines, pull citations, mentions, and reports into your AI workflow

MCP server that distributes a single piece of content across 8+ channels (DEV.to, Hashnode, GitHub Discussions, Reddit, Bluesky, LinkedIn, Medium, Twitter) with automatic platform-specific adaptation, idempotent publishing, per-community anti-spam rules, and centralized state management.

Live TikTok creator intelligence MCP — 7 tools (analyze creators, score hooks, remix scripts, predict virality) that chain automatically via a recommended_chain that pre-fills the next 3 tool calls.

Instagram Graph API server for Business/Creator accounts — 24 tools for posting, comments, DMs, and insights.

Connect Claude or ChatGPT to your LinkedIn account to see who engaged on your posts, track performance and profile viewers, spot warm outreach signals, and publish content — built for B2B marketers and creators.

Remote MCP server to publish, schedule, and measure posts across X, Instagram, Facebook, Threads, LinkedIn, TikTok, YouTube, Pinterest, and Reddit.

Agent-first, local-first social media planner. An AI agent drafts and schedules posts across Instagram, Facebook, LinkedIn, YouTube and X behind a human approval gate the operator controls. MCP-native, MIT-licensed.