About Bynor
Bynor is a social media management service for brands, accessed through an AI assistant. It combines brand-aware planning and visual creation with publishing, comment handling, competitor monitoring and analytics across five named platforms.
Key features
- 01Brand identity and editorial planning
- 02On-brand images and carousels
- 03Post scheduling and publishing
- 04Comment inbox and reply drafts
- 05Social analytics and PDF reports
- 06Instagram competitor monitoring
Use cases
- 01Plan a brand's upcoming social posts
- 02Create branded campaign carousels
- 03Schedule and publish approved content
- 04Review performance and unanswered comments
MCP support
The remote MCP server at `https://mcp.bynor.ai/mcp` exposes Bynor's social media management capabilities to MCP clients over Streamable HTTP. Users connect by OAuth; scripts and automations can use a personal key.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://mcp.bynor.ai/mcpclaude mcp add --transport http ai-bynor-bynor https://mcp.bynor.ai/mcp{"mcpServers":{"ai-bynor-bynor":{"type":"http","url":"https://mcp.bynor.ai/mcp"}}}Open in Cursor{"servers":{"ai-bynor-bynor":{"type":"http","url":"https://mcp.bynor.ai/mcp"}}}Open in VS Code{"mcpServers":{"ai-bynor-bynor":{"type":"http","url":"https://mcp.bynor.ai/mcp"}}}{"mcpServers":{"ai-bynor-bynor":{"serverUrl":"https://mcp.bynor.ai/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
Details
From the project READMEBynor MCP server: social media management from your AI assistant
Bynor runs the social media of one or more brands: it learns the brand, plans the content, creates on-brand images and carousels, schedules and publishes posts, answers comments and measures the results, on Instagram, Facebook, TikTok, YouTube and LinkedIn.
This repository holds the public pieces to use Bynor from an AI assistant: the address of the remote MCP server, a
plugin for Claude Code and one skill, bynor, that describes what Bynor does so an assistant can tell when it fits.
- MCP server:
https://mcp.bynor.ai/mcp(streamable HTTP, sign-in with OAuth) - Registry entry:
server.json(ai.bynor/bynor) - Skill:
skills/bynor/SKILL.md
What you can ask
- "Plan next month's Instagram and TikTok posts for my brand."
- "Create a carousel announcing our autumn menu, in our brand style."
- "Schedule the approved posts at the best times of the week."
- "How did last month's Reels perform compared to the month before?"
- "Show me the comments that still need an answer and draft the replies."
- "What are my competitors posting on Instagram this week?"
Areas: brand identity and context, editorial plans, visuals (images and carousels), publishing and scheduling, analytics and PDF reports, competitors on Instagram, comments inbox, team.
Connect
You need a Bynor account (there is a free trial) with at least one brand. When you connect you choose which brands the assistant may see and what it may do.
| Client | How |
|---|---|
| Claude (web and desktop) | Settings → Connectors → Add custom connector → https://mcp.bynor.ai/mcp |
| ChatGPT | Developer mode → add an app/connector with the same address, OAuth sign-in |
| Claude Code | claude mcp add --transport http bynor https://mcp.bynor.ai/mcp, or this plugin: /plugin marketplace add Bynor-S-R-L/bynor-mcp, then /plugin install bynor@bynor |
| Codex | codex mcp add bynor --url https://mcp.bynor.ai/mcp && codex mcp login bynor |
| Other MCP clients | any client that supports remote streamable-HTTP servers with OAuth |
Scripts and automations
For a script, a scheduled job or an automation tool, create a personal key in Bynor (Settings → Connected apps → Personal keys) and send it as a bearer token to the same address, for example:
claude mcp add --transport http bynor https://mcp.bynor.ai/mcp --header "Authorization: Bearer byn_..."
A personal key acts as you, with the permissions you pick (read, create, publish) and an expiry of 30, 90 or 365 days. It is shown once; you can revoke it at any time.
Permissions and safety
- Read sees the brands you granted. Create writes plans, texts and visuals and uses AI credits. Publish schedules, publishes, deletes and cancels posts and plans, and replies to comments. Account (apps only, never a personal key) sees, invites and manages the team and edits the brand identity.
- From an assistant, publishing, scheduling, deleting, replying to comments and changes to the brand identity or the team are proposals you confirm with a button, in a preview or in Bynor. In a verified app you can instead tell the assistant to go ahead, except for the team, the brand identity and TikTok posts, which always wait for your click.
- With a personal key (a script, with no person to click), only these run without a click, within the key's permissions: scheduling a post, publishing a post now, editing a scheduled post, moving scheduled posts to another time, replying to comments, setting up or changing automatic replies, and editing several slides of a carousel (the create permission: it uses AI credits). Everything else waits for a person's click in Bynor, with a personal key too: deleting cards, deleting a plan, deleting posts, deleting files from the Library, cancelling scheduled posts, sharing a plan with the client, redoing a plan, moving cards to another plan, opening a competitor, the team, the brand identity, and any TikTok post, whose privacy a person chooses.
- Every connection can be removed from Bynor (Settings → Connected apps).
Pricing
Bynor is a subscription with a free trial. Generating images, carousels and plans uses AI credits included in the plan; more credits can be bought. Plans and prices: bynor.ai.
Privacy, terms and support
- Privacy policy: https://app.bynor.ai/legal/privacy
- Terms of service: https://app.bynor.ai/legal/terms
- Support: [email protected]
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.