The full capability story

Everything you need to publish with control

Multi-network publishing, compose-once-adapt-per-network, approval gates with an immutable audit log, assistive AI, scheduling, analytics, and team RBAC — all reachable from the dashboard, the CLI, the REST API, or an MCP tool.

Publishing

Multi-network publishing without the per-channel tax

A profile is a cheap Postgres primitive you can create over the API. Manage many brand or client accounts across every major network from one control plane.

  • 12 networks: X, LinkedIn, Instagram, Facebook, Threads, Reddit, YouTube, TikTok, Pinterest, Google Business, Bluesky, Mastodon
  • Per-network versioned adapters — behaviour driven server-side, not hardcoded in the client
  • Connect via OAuth redirect or an app-password / handle form, depending on the network

Compose

Compose once, adapt per network

One source asset becomes platform-native variants — length, hashtags, mentions, and threadified chains — with each network's constraints enforced before anything ships.

  • Length, media, and format limits validated per network
  • Automatic threadify when content exceeds a network's character cap
  • Per-network overrides when you want a variant to differ on purpose

Governance

Approval gates + an immutable audit log

Drafts require human sign-off before anything reaches a connected account. Every write records actor, content, target, and result — append-only, never edited.

  • Mandatory approval gate on every outbound write
  • Immutable audit trail: who, what, where, and the provider result
  • Scoped API keys so each token only does what it should

AI

Assistive AI, always behind the gate

Per-network reformatting and threadify are powered by Claude — assistive, never autonomous. The AI proposes; a human still approves before publish.

  • Reshape a source post into each network's native form
  • Suggested hashtags and threading, surfaced as a draft
  • Every AI-assisted draft still passes through the same approval gate

Timing

Scheduling & a durable queue

Schedule posts to publish at a precise time, or queue them to flow out on a cadence. A durable scheduler loop drives delivery and records the outcome.

  • Schedule at an exact UTC timestamp
  • Durable worker loop with retries and recorded results
  • Queue health, last error, and token status readable over the API

Insight

Analytics you can query

Token health, rate-limit headroom, last error, and per-profile engagement — all readable over the API and MCP, not just trapped in a dashboard.

  • Per-profile engagement and delivery metrics
  • Rate-limit headroom and token expiry surfaced before they bite
  • Everything available as JSON over the API and as MCP tools

Teams

Team & role-based access control

Invite teammates, assign roles, and keep tenants isolated at the database layer with Postgres row-level security — not just an application WHERE clause.

  • Invite seats and assign roles per workspace
  • Postgres row-level security enforces tenant isolation in the database
  • Audit log attributes every action to a specific actor

Interfaces

API + CLI + MCP on every tier

The API and CLI are the product. An MCP server exposes every verb as a tool, so agent fleets can operate profiles with the exact same parity as the dashboard.

  • REST API at /api/v1 — OpenAPI documented, Bearer API-key auth
  • Flagship CLI mirrors the API 1:1, --json everywhere, --dry-run on writes
  • MCP server — every verb as a tool, on every plan including Free

Publishes to 12 networks

Each has its own landing page with connect steps and an example.

Ship with control

Start free, drive it from the API and CLI, and keep every AI action behind a human gate.