Skip to main content
Glama

Ceptory MCP

Connect any MCP agent (Claude, Cursor, Codex, Gemini CLI, VS Code, Windsurf, the OpenAI Agents SDK, LangChain…) to Ceptory. In plain language, the agent can:

  • set up camera workflows (PPE, person down, fire and hazards, intrusion, face access, loitering, unattended bags, crowd limits, illegal parking, rider helmets)

  • change rules, schedules, zones and Discord alerts, and pause or resume cameras

  • read, acknowledge and resolve alerts, with their photos

  • group cameras into sites and follow people and vehicles across cameras

  • upload video, search it in plain language, and ask questions about it

  • run any single AI detection engine on an image

The agent acts as you. It uses your workspace, role and prepaid balance, with the same checks as the Ceptory dashboard.

1. Get a token

Go to ceptory.com/dashboard/mcp → Create token. The token starts with cpd_ and is shown once. You can revoke it there at any time.

Related MCP server: ELEMS MCP

2. Connect your agent

The hosted endpoint is https://www.ceptory.com/api/mcp. It uses Streamable HTTP with MCP 2026-07-28, and 2025-era clients also work. Authenticate with Authorization: Bearer <token>.

Claude Code

claude mcp add --transport http ceptory https://www.ceptory.com/api/mcp \
  --header "Authorization: Bearer <token>"

Claude Desktop: add this to claude_desktop_config.json. It uses this package as a stdio bridge and needs Node.js 20 or later.

{ "mcpServers": { "ceptory": { "command": "npx", "args": ["-y", "ceptory-mcp"], "env": { "CEPTORY_TOKEN": "<token>" } } } }

Cursor: add this to ~/.cursor/mcp.json.

{ "mcpServers": { "ceptory": { "url": "https://www.ceptory.com/api/mcp", "headers": { "Authorization": "Bearer <token>" } } } }

VS Code: add this to .vscode/mcp.json.

{ "servers": { "ceptory": { "type": "http", "url": "https://www.ceptory.com/api/mcp", "headers": { "Authorization": "Bearer <token>" } } } }

Codex: add this to ~/.codex/config.toml.

[mcp_servers.ceptory]
command = "npx"
args = ["-y", "ceptory-mcp"]
env = { CEPTORY_TOKEN = "<token>" }

Gemini CLI: add this to ~/.gemini/settings.json.

{ "mcpServers": { "ceptory": { "httpUrl": "https://www.ceptory.com/api/mcp", "headers": { "Authorization": "Bearer <token>" } } } }

Anything else: point any MCP client at the URL with the header, or run CEPTORY_TOKEN=<token> npx -y ceptory-mcp for stdio.

3. Ask

  • "Set up a camera called Loading dock that alerts our Discord when anyone isn't wearing a helmet and a hi-vis vest."

  • "Watch the back fence for people or vehicles between 10pm and 6am India time."

  • "Which cameras are offline? Show me today's open alerts with photos."

  • "Group the gate and lobby cameras as North site and show me where people went in the last hour."

  • "Make a chat called Night shift, upload ~/Downloads/gate.mp4, and tell me when a white van appears."

The full tool reference is at ceptory.com/mcp.

What this package does

The tools run on Ceptory. This package provides two things:

  1. A stdio bridge (ceptory-mcp). It connects stdio-only clients to the hosted server. It also adds what only a local process can do:

    • upload_video_file: uploads a video from your computer (up to 2 GB) and starts indexing it.

    • image_path: tools that take an image accept a local JPEG or PNG path as well as base64.

  2. An HTTP gateway (ceptory-mcp --http). It runs a standalone /mcp endpoint, for example on Railway, and forwards to Ceptory. It holds no credentials: each client sends its own token. GET /health reports health.

Variable

Default

Used by

CEPTORY_TOKEN

(none)

stdio bridge (required)

CEPTORY_MCP_URL

https://www.ceptory.com/api/mcp

both

PORT / HOST

3333 / 0.0.0.0

gateway

Develop

npm install
npm run build
npx @modelcontextprotocol/inspector --cli -e CEPTORY_TOKEN=cpd_… node build/index.js --method tools/list
npx @modelcontextprotocol/inspector --cli -e CEPTORY_TOKEN=cpd_… node build/index.js --protocol-era modern --method tools/list

Logs go to stderr. In stdio mode, stdout carries JSON-RPC only.

License

MIT

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      B
      maintenance
      Enables agent clients to safely connect to tools and execution resources through MCP with authorization, approvals, audit, chat-context isolation, SSH/Docker access, and long-running command session tracking.
      MIT
    • F
      license
      Not graded
      quality
      B
      maintenance
      Connects MCP-capable agents to an authenticated ELEMS account, enabling website creation and inventory management, page and element editing, media and template handling, structured content authoring, draft previews, and explicit publishing.
      -
    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables an agent to search approved MCP integrations and expose only the tools a given task needs, signing the user in once via browser-based OAuth and enforcing scopes, effects and budgets in code. It also acts as a single gateway connection so existing hosts can reach many servers with per-tenant isolation and audit events.
      MIT
    • A
      license
      Not graded
      quality
      B
      maintenance
      Connects AI agents and MCP clients to a hosted gateway for authenticated tool discovery and governed execution, including inspecting campaigns and creating empty campaigns for manual configuration.
      MIT