# Caelario [Health: Active]

**Category:** 🌐 Social Media  
**Repository:** https://github.com/chunkydonut21/caelario-mcp  
**GitHub Stars:** 0  
**npm Downloads (last month):** 1033  
**Views:** 1  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/caelario-2

## Description
MCP server for Instagram, TikTok, YouTube, X and Facebook data extraction and automation.

## Tools
Capabilities this server exposes over MCP:

- **discover_platforms** — List every Caelario platform, operation, limit, and price.
- **get_operation_guide** — Explain and validate one operation before spending credits.
- **get_account** — Return workspace identity and extraction limits.
- **get_credit_usage** — Return available, reserved, promotional, purchased, and spent credits.
- **estimate_credits** — Estimate the maximum reservation for one or many targets before starting work.
- **extract_now** — Run one bounded target synchronously and return a model-safe result page.
- **create_extraction** — Create one durable extraction job for work that may continue in the background.
- **list_extractions** — List recent durable extraction jobs without embedding their large results.
- **get_extraction** — Get one job's progress, billing, and error metadata.
- **get_extraction_result** — Page through a completed job's structured result without flooding model context.
- **wait_for_extraction** — Wait briefly for a durable job to become terminal without returning its large result.
- **cancel_extraction** — Cancel a queued job or request cancellation of a running job.
- **create_batch** — Run one operation for up to 100 targets as an expandable durable batch.
- **list_batches** — List recent batches and optionally their child jobs.
- **get_batch** — Get aggregate status and child jobs for one batch.
- **cancel_batch** — Cancel queued batch children and request cancellation of running children.
- **list_schedules** — List active and paused extraction schedules.
- **get_schedule** — Get one schedule and its recent run history.
- **create_schedule** — Create a timezone-aware once, daily, weekly, or monthly extraction schedule.
- **update_schedule** — Replace a schedule definition while preserving its run history.
- **manage_schedule** — Pause, resume, run now, or delete an extraction schedule.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "caelario": {
    "command": "npx",
    "args": ["-y","caelario-mcp@latest"]
  }
}
```

## Documentation

## What Caelario does

Caelario provides MCP access to social-media data workflows covering Instagram, TikTok, YouTube, X, and Facebook. It supports capability discovery, extraction requests, usage tracking, result retrieval, batch processing, and recurring schedules.

The tool set separates short-running work from jobs that may continue in the background. `extract_now` handles one bounded target synchronously, while durable jobs can be created, monitored, cancelled, and read through separate operations. Completed results can be paged so large responses do not fill the model context.

Credit-aware operations are built into the workflow. Agents can inspect workspace identity and limits, review credit balances, learn an operation’s requirements, and estimate the maximum reservation before starting work.

## How it works

The local package registers a stdio MCP server with the selected client. Requests are sent to Caelario’s hosted service, which manages provider identities and the underlying acquisition infrastructure. The installer does not request separate Instagram, TikTok, YouTube, X, or Facebook credentials.

For longer work, an agent can create an extraction job and then use listing, status, waiting, cancellation, and result-pagination tools. Batches support up to 100 targets for one operation and expose aggregate status along with child jobs. Schedules can run once, daily, weekly, or monthly, with timezone-aware definitions and controls to pause, resume, run immediately, update, or delete them.

If a request times out, the returned `idempotency_key` can be reused with identical arguments to recover the original request rather than initiate another paid extraction. The default HTTP timeout is 120 seconds. Failed jobs retain their error information and remain marked as not ready.

## Setup and configuration

The guided setup is available at [app.caelario.com/connect](https://app.caelario.com/connect). Create a Caelario account, choose Codex or Claude Code, and generate a workspace connection key. New accounts start with 200 free credits.

Node.js 20 or newer is required. Install for Codex with:

```bash
npx -y caelario-mcp@latest install codex
```

Install for Claude Code with:

```bash
npx -y caelario-mcp@latest install claude
```

To configure Claude Desktop explicitly, use:

```bash
npx -y caelario-mcp@latest install claude-desktop
```

The installer registers the server at user scope and preserves existing desktop settings and other MCP servers. On first setup, or when a key is rejected, it opens the guided setup and securely prompts for a replacement key in the terminal. Restart the client after installation and approve Caelario when prompted.

## Tools and capabilities

- Discover available platforms, operations, limits, and prices.
- Validate an operation and estimate credit reservations before extraction.
- Inspect workspace identity, extraction limits, and credit categories.
- Run bounded synchronous extractions.
- Create, list, inspect, wait for, cancel, and paginate durable extraction jobs.
- Create and manage batches covering up to 100 targets.
- Create and control timezone-aware recurring extraction schedules.

## Limitations and notes

The connector is MIT licensed, but Caelario’s hosted API, provider implementations, acquisition infrastructure, and managed identities are proprietary services outside the package. Extraction work uses credits, including reservations associated with operations. Large job results should be fetched through pagination rather than embedded in job listings or status responses.

The package includes MCP Registry metadata under `io.github.chunkydonut21/caelario`. npm publication precedes the separate Registry publication step.

_Full upstream README: https://allmcps.com/mcp/caelario-2/readme_