# Magic Hour MCP

> Magic Hour MCP is a hosted server with 50 tools that create and edit images and video, from text-to-video, lip sync and face swap to upscaling, subtitles and music, billed in Magic Hour credits.

- Canonical: https://www.anchorterminal.com/tools/magic-hour-mcp
- Pricing: https://www.anchorterminal.com/tools/magic-hour-mcp#pricing
- Reviews: https://www.anchorterminal.com/tools/magic-hour-mcp#reviews
- Markdown: https://www.anchorterminal.com/tools/magic-hour-mcp.md (~4,750 tokens)
- Slim: https://www.anchorterminal.com/tools/magic-hour-mcp.min.md (~1,280 tokens, same facts, less prose, for token-sensitive contexts)
- JSON: https://www.anchorterminal.com/tools/magic-hour-mcp.json (this page as data, same URL with Accept: application/json)
- Site index for agents: https://www.anchorterminal.com/llms.txt (the key docs and every listing in one file, at most 16 MB, about 4 million tokens: https://www.anchorterminal.com/llms-full.txt)
- API: https://www.anchorterminal.com/api/v1/index.json
- MCP: https://www.anchorterminal.com/mcp (the whole catalogue as tools, no key)
- DNS: dig +short TXT magic-hour-mcp.grade.anchorterminal.com (the grade as a DNS TXT record, for a sandbox with no HTTP where names still resolve; also top., best.<category>. and <a>.vs.<b>. under the zone, and the same under grade.letme.dev)
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)
- Updated: 2026-11-08

## Overview

**Grade D · 50.7/100 · rank #1053 of 1510 · #11 in Video generation · not agent-ready · confidence medium**


## Assessment

Magic Hour's MCP server exposes its whole image and video toolkit, lip sync, face swap, upscaling and text-to-video, with typed schemas built from its OpenAPI file and every tool annotated. At 50 tools and 160 KB of definitions it's heavy, rate limits aren't published, and API access needs a paid plan.

## Facts

| Field | Value |
| --- | --- |
| Vendor | Magic Hour AI, Inc. (https://magichour.ai) |
| Kind | MCP server |
| Category | Video generation (https://www.anchorterminal.com/categories/video-generation) |
| Transport | Streamable HTTP |
| Endpoint | `https://mcp.magichour.ai/` |
| Auth | OAuth or key · Tool discovery is public. Calls need a Magic Hour API key as a Bearer token (Claude Code, Codex CLI, Manus), or OAuth sign-in through the Claude and ChatGPT connectors. |
| Pricing | Freemium (Freemium) · Paid plans from $12 a month billed yearly ($19 monthly) with 144,000 credits a year and API access; one-time credit packs without a subscription. A free account gets 400 sign-up credits, and some public tools work without an account; API access is listed on paid plans (checked 2026-11-07). |
| x402 | No · No x402, MPP or L402 on the MCP docs or pricing page (checked 2026-11-07). |
| Licence | unknown |
| Tools exposed | 50 |
| MCP registry name | `io.github.magichourhq/magic-hour` |
| Source | https://github.com/magichourhq/magic-hour-mcp |
| Docs | https://docs.magichour.ai/integration/model-context-protocol |
| llms.txt | https://docs.magichour.ai/llms.txt |
| Surface listed | The hosted MCP server at mcp.magichour.ai, read from docs.magichour.ai, magichour.ai and its tools/list on 7 November 2026. No tool was called |
| Tools | Video (text, image or audio to video, lip sync, translate, restyle, subtitles, upscale, face and character swap), images (generate, edit, headshots, background removal, upscale, face and head swap), music, uploads, project status and downloads |
| Capabilities | video.generate, video.image-to-video, video.edit, image.generate, image.edit, image.upscale, music.generate |
| Tags | official, hosted, mcp, remote-mcp, api-key, oauth, freemium, status-page, llms-txt, mcp-registry, annotated-tools |
| JSON | https://www.anchorterminal.com/api/v1/tools/magic-hour-mcp.json |

## Score breakdown (methodology v0.7, October 2026 research run)

Assessed 2026-11-07 from public evidence against the published checklist (https://www.anchorterminal.com/benchmark/#checklist). Confidence: medium. Performance and Task success pending (no score, not in the total); the total is Σ(score × weight) ÷ 80 over the 7 assessed categories. "This run" is each category's share of the 100 points.

| Category | Weight | This run | Score (0–100) | Points |
| --- | --- | --- | --- | --- |
| Reliability | 16% | 20 | 35 | 7.0 |
| Performance | 10% | pending | pending | n/a |
| Schema & documentation | 13% | 16.2 | 82 | 13.3 |
| Agent ergonomics | 13% | 16.2 | 55 | 8.9 |
| Security & auth | 14% | 17.5 | 45 | 7.9 |
| Payments & pricing | 10% | 12.5 | 15 | 1.9 |
| Task success | 10% | pending | pending | n/a |
| Maintenance & community | 7% | 8.8 | 75 | 6.6 |
| Transparency & trust (editorial 50, provenance 65) | 7% | 8.8 | 58 | 5.1 |
| Negative events | up to −15 | up to −15 | none recorded | 0 |
| **Total** | | | | **50.7 → D** |

### Why each score

- Reliability 35 (checked 2026-11-07): Status page at status.magichour.ai, all services online, with maintenance and previous incidents sections; no MCP component read (15). Incident history not read (5). No API rate limits found (0). No retry guidance found; wait_for_* tools poll long jobs (5 of 15). No SLA found (0). Not labelled beta (10). 15 + 5 + 0 + 5 + 0 + 10 = 35.
- Performance: Pending. Latency is measured per call by our probes, which haven't run yet, so this run doesn't score it. Its weight is shared across the assessed categories until the first probe window closes.
- Schema & documentation 82 (checked 2026-11-07): All 50 tools read from tools/list with JSON Schema inputs and 29 enums, built from Magic Hour's OpenAPI specification (25). llms.txt (10). Long descriptions with a get_workflow_guide tool for choosing between them (17 of 20). Typed inputs with enums and no free-form objects (15). Worked examples; error responses not documented (5 of 15). Dated changelog with entries on 3 to 5 November 2026 (10 of 15, product changelog rather than MCP versions). 25 + 10 + 17 + 15 + 5 + 10 = 82.
- Agent ergonomics 55 (checked 2026-11-07): 50 tools totalling about 160 KB of definitions (5). list_saved_items and project getters; limited size controls (10 of 20). Error handling not documented (5 of 20). Every tool annotated, 14 read-only and the three deletes marked destructive (20). Few required inputs and SDKs (Python, Node and others listed in the docs) (15). 5 + 10 + 5 + 20 + 15 = 55.
- Security & auth 45 (checked 2026-11-07): API keys as Bearer tokens or OAuth through the Claude and ChatGPT connectors (20 of 30). Destructive deletes are annotated; no read-only key (8 of 20). Tools return generated media, not third-party text (10). No audit log found (0). No security.txt or certifications found; face and voice tools come with policy limits in the terms (7 of 20). 20 + 8 + 10 + 0 + 7 = 45.
- Payments & pricing 15 (checked 2026-11-07): No machine payment protocol (0). Plan and credit-pack prices public, per-tool credit costs not read (10). 400 sign-up credits with no card, but API access sits on paid plans (5 of 20). A person signs up for a key in a browser (0). Tool discovery is open, but calls need a key (0). 0 + 10 + 5 + 0 = 15.
- Task success: Pending. Task success needs the category task suites run through each tool, which haven't run yet, so this run doesn't score it. Its weight is shared across the assessed categories until then. A data provider's data-quality score is published on its listing now and becomes half of this category when it's scored.
- Maintenance & community 75 (checked 2026-11-07): Changelog entries on 3, 4 and 5 November 2026 (30). More than three dated entries in 90 days (20). Changelog and support; issues on the public MCP repository not read (10 of 25). Registry entry io.github.magichourhq/magic-hour (0.1.1, 4 November) (15). No package health read (0). 30 + 20 + 10 + 15 + 0 = 75. Registry presence read from Anchor Terminal's own daily import of the official MCP registry (/api/v1/registry.json, 6 November 2026), not the registry's API.
- Transparency & trust 58 (checked 2026-11-07): Closed service under terms naming Magic Hour AI, Inc., revised 3 October 2026 (15). The privacy policy says prompts, uploads and outputs aren't sold or used to train models, for free and paid accounts, and MCP requests are handled the same way; retention periods aren't given (20 of 30). No deprecation policy found (5 of 20). Subprocessors not read (10 of 20, model providers named in tool docs). 15 + 20 + 5 + 10 = 50.

Fix list, everything this grade says the listing lacks, the biggest gain first (14 items). What it covers is at https://www.anchorterminal.com/fixes/magic-hour-mcp.md (JSON https://www.anchorterminal.com/fixes/magic-hour-mcp.json). The whole list for a coding agent comes from `POST https://www.anchorterminal.com/api/v1/fix-list` with `{"listing": "magic-hour-mcp", "plan": "read"}`, free while the beta runs

### What we couldn't check

- Per-tool credit costs not read
- Incident history not read
- People, ownership and investors not researched

### Sources

- MCP docs: <https://docs.magichour.ai/integration/model-context-protocol> (seen 2026-11-07)
- tools/list, 50 tools: <https://mcp.magichour.ai/> (seen 2026-11-07)
- pricing: <https://magichour.ai/pricing> (seen 2026-11-07)
- changelog: <https://docs.magichour.ai/changelog> (seen 2026-11-07)
- terms and privacy: <https://magichour.ai/terms-of-service> (seen 2026-11-07)
- status page: <https://status.magichour.ai> (seen 2026-11-07)
- lead, awesome-remote-mcp-servers pull request #2665: <https://github.com/punkpeye/awesome-remote-mcp-servers/pull/2665> (seen 2026-11-07)

## Score history

Every move of a score since the listing was graded, newest first, with its cause. A score can't move without an entry here. JSON: https://www.anchorterminal.com/api/v1/tools/magic-hour-mcp/history.json

- 2026-11-07: graded 50.7 D (methodology 0.7)

## Who's behind it (provenance 65/100, checked 2026-11-07)

| Check | Finding | Points |
| --- | --- | --- |
| Legal entity named | Magic Hour AI, Inc. | 20/20 |
| Domain age | magichour.ai, no registry record we could read | 0/15 |
| Endpoint on the vendor's domain | mcp.magichour.ai | 15/15 |
| Terms of service | published, not read by us yet | 4/10 |
| Privacy policy | published, not read by us yet | 4/10 |
| Status page | status.magichour.ai | 10/10 |
| Changelog | published | 10/10 |
| security.txt | not found | 0/10 |
| People named | not researched yet | n/a |
| Support route | help pages | 5/5 |

Not looked up: RDAP isn't requested under the 1 November 2026 rules.

Terms revised 3 October 2026 naming Magic Hour AI, Inc.

### Terms and privacy, as read

A reading by a fixed set of rules, each answered with the vendor's own sentence. Not legal advice.

**Terms of service** (https://magichour.ai/terms-of-service), not read yet.


**Privacy policy** (https://magichour.ai/privacy-policy), not read yet.


## Humans behind the project (researched 2026-11-07 from public pages, not part of the grade)

### Key people

- No people named yet.
- Note: People, ownership and investors weren't researched in this batch (7 November 2026, leads from the awesome-remote-mcp-servers pull requests).

### Getting support

- Help: https://docs.magichour.ai/integration/model-context-protocol
- Sources: https://docs.magichour.ai/integration/model-context-protocol

## Live (updated 2026-11-08 00:43 UTC)

- Right now: up, HTTP 200, 152 ms, checked 2026-11-08 00:43 UTC (mcp-initialize on `https://mcp.magichour.ai/`)
- Uptime 24h 100.0% (61 probes) · 30 days 100.0% (61 probes) · p50 165 ms · p95 240 ms
- Vendor status page: not read, no machine-readable status found
- Always current: https://www.anchorterminal.com/api/v1/live/magic-hour-mcp.json

## Probe metrics

Not measured yet. Our benchmark probes haven't run, so there's no availability, latency or error rate from a run and Performance is pending. Live uptime, where we poll the endpoint, is under Live and doesn't change the score.

## Strengths

- 50 tools with typed JSON Schema built from the OpenAPI specification
- Every tool annotated; deletes marked destructive
- A workflow guide and wait tools for long generation jobs
- Prompts, uploads and outputs aren't used for training
- Tool discovery works without a key

## Weaknesses

- About 160 KB of tool definitions
- No published rate limits or error docs
- API access sits on paid plans
- Face swap and head swap tools need care with consent



## Before you call it (notes for agents)

1. Call get_workflow_guide first to pick the right tool
2. Use the wait_for_* tools instead of polling project status yourself
3. Check credits with get_account_info before long video jobs
4. Only swap faces or voices of people who consented

## Connect

Claude Code:

```bash
claude mcp add magic-hour --transport http https://mcp.magichour.ai/ --header "Authorization: Bearer $MAGIC_HOUR_API_KEY"
```

MCP client configuration:

```json
{
  "mcpServers": {
    "magic-hour": {
      "headers": {
        "Authorization": "Bearer YOUR_MAGIC_HOUR_API_KEY"
      },
      "url": "https://mcp.magichour.ai/"
    }
  }
}
```

Through letme: https://letme.dev/magic-hour-mcp (letme picks it for music.generate, the top-graded tool for the job). letme answers with the pick and how to call it direct. How it works: https://www.anchorterminal.com/letme/index.md

## Similar tools

Ranked by shared capabilities, then score. Same-category tools with no shared capability key are listed last.

| Tool | Grade | Score | Rank | Shared capabilities | x402 | Markdown |
| --- | --- | --- | --- | --- | --- | --- |
| Higgsfield API | B | 62.7 | 542 | video.generate, video.image-to-video, video.edit, image.generate, image.edit | no | https://www.anchorterminal.com/tools/higgsfield.md |
| FLORA | B | 62.6 | 552 | image.generate, image.edit, image.upscale, video.generate, video.image-to-video | no | https://www.anchorterminal.com/tools/flora.md |
| Ideogram API | D | 52.2 | 997 | image.generate, image.edit, image.upscale, video.generate, video.image-to-video | no | https://www.anchorterminal.com/tools/ideogram.md |
| Melius | D | 53.3 | 954 | image.generate, image.edit, image.upscale, video.generate | no | https://www.anchorterminal.com/tools/melius.md |
| Osmo CLI | F | 35.2 | 1397 | image.edit, video.generate, video.image-to-video, image.generate | no | https://www.anchorterminal.com/tools/osmo-cli.md |
| fal video models | BB | 71 | 174 | video.generate, video.image-to-video, video.edit | no | https://www.anchorterminal.com/tools/fal-video.md |

## Panel reviews (0)

Reviewed by the Anchor panel (https://www.anchorterminal.com/reviewers/index.md): .

Desk reviews, written from public documentation, pricing, terms, source and status history on 1 October 2026. No calls made. For a desk review, the outcome says whether the reviewer's questions could be answered from public material: success, partial or failure. How reviews work: https://www.anchorterminal.com/reviews/how-it-works.md

## Notable

- Tools are built from the bundled Magic Hour OpenAPI specification plus helpers (source: <https://docs.magichour.ai/integration/model-context-protocol>)
- Discovery is public; tool calls need an API key or OAuth (source: <https://docs.magichour.ai/integration/model-context-protocol>)
- No-training policy for free and paid accounts, MCP included (source: <https://magichour.ai/privacy-policy>)
- Paid plans from $12 a month billed yearly with API access (source: <https://magichour.ai/pricing>)

- #11 of 19 in Best video generation for AI agents: https://www.anchorterminal.com/best/video-generation/index.md
- All 183 video comparisons: https://www.anchorterminal.com/compare/video-generation/index.md

## Compare

- [Alibaba Wan (Model Studio) vs Magic Hour MCP](https://www.anchorterminal.com/compare/alibaba-wan-vs-magic-hour-mcp.md): B 65.2 vs D 50.7
- [fal video models vs Magic Hour MCP](https://www.anchorterminal.com/compare/fal-video-vs-magic-hour-mcp.md): BB 71 vs D 50.7
- [framefields vs Magic Hour MCP](https://www.anchorterminal.com/compare/framefields-vs-magic-hour-mcp.md): D 53.9 vs D 50.7
- [Google Veo vs Magic Hour MCP](https://www.anchorterminal.com/compare/google-veo-vs-magic-hour-mcp.md): B 69.3 vs D 50.7
- [Higgsfield API vs Magic Hour MCP](https://www.anchorterminal.com/compare/higgsfield-vs-magic-hour-mcp.md): B 62.7 vs D 50.7
- [HyperFrames vs Magic Hour MCP](https://www.anchorterminal.com/compare/hyperframes-vs-magic-hour-mcp.md): E 39.7 vs D 50.7
- [Kling AI API vs Magic Hour MCP](https://www.anchorterminal.com/compare/kling-vs-magic-hour-mcp.md): F 22 vs D 50.7
- [Luma AI API vs Magic Hour MCP](https://www.anchorterminal.com/compare/luma-vs-magic-hour-mcp.md): D 50.5 vs D 50.7
- [Magic Hour MCP vs MiniMax Video API](https://www.anchorterminal.com/compare/magic-hour-mcp-vs-minimax-video.md): D 50.7 vs B 62
- [Magic Hour MCP vs OpenAI Sora API](https://www.anchorterminal.com/compare/magic-hour-mcp-vs-openai-sora.md): D 50.7 vs F 9.4
- [Magic Hour MCP vs Pika API](https://www.anchorterminal.com/compare/magic-hour-mcp-vs-pika.md): D 50.7 vs E 40.5
- [Magic Hour MCP vs PixVerse API](https://www.anchorterminal.com/compare/magic-hour-mcp-vs-pixverse.md): D 50.7 vs E 42.5
- [Magic Hour MCP vs Popcraft MCP](https://www.anchorterminal.com/compare/magic-hour-mcp-vs-popcraft-mcp.md): D 50.7 vs E 38
- [Magic Hour MCP vs Replicate video models](https://www.anchorterminal.com/compare/magic-hour-mcp-vs-replicate-video.md): D 50.7 vs B 64.2
- [Magic Hour MCP vs Runway API](https://www.anchorterminal.com/compare/magic-hour-mcp-vs-runway.md): D 50.7 vs C 59.2
- [Magic Hour MCP vs Seedance on BytePlus ModelArk](https://www.anchorterminal.com/compare/magic-hour-mcp-vs-seedance-on-byteplus-modelark.md): D 50.7 vs C 59.4
- [Magic Hour MCP vs Vidu API](https://www.anchorterminal.com/compare/magic-hour-mcp-vs-vidu.md): D 50.7 vs E 40.2
- [Clueso vs Magic Hour MCP](https://www.anchorterminal.com/compare/clueso-video-vs-magic-hour-mcp.md): F 36.3 vs D 50.7
- [Magic Hour MCP vs sync. labs MCP server](https://www.anchorterminal.com/compare/magic-hour-mcp-vs-sync-labs-mcp.md): D 50.7 vs D 51.5

## Verify this listing

For the vendor. The badge or a plain link to this page verifies the listing, from a page on magichour.ai or one of its subdomains, or the README at the root of github.com/magichourhq/magic-hour-mcp. It shows the listing is the vendor's and that the vendor knows it's here, and it never changes a grade, rank or review. The vendor sends the page's address to `POST https://www.anchorterminal.com/api/v1/verify` as `{"slug": "magic-hour-mcp", "url": "…"}`, or calls the `verify_listing` tool at https://www.anchorterminal.com/mcp. We fetch the page once, then again every week; two failed checks in a row and the verification lapses, and a later pass restores it. What we check: https://www.anchorterminal.com/builders/index.md#verify

HTML badge:

```html
<a href="https://www.anchorterminal.com/tools/magic-hour-mcp"><img src="https://www.anchorterminal.com/badges/magic-hour-mcp.svg" alt="Magic Hour MCP on Anchor Terminal" height="20"></a>
```

Markdown badge, for a README:

```markdown
[![Magic Hour MCP on Anchor Terminal](https://www.anchorterminal.com/badges/magic-hour-mcp.svg)](https://www.anchorterminal.com/tools/magic-hour-mcp)
```

Plain link:

```html
<a href="https://www.anchorterminal.com/tools/magic-hour-mcp">Magic Hour MCP on Anchor Terminal</a>
```

## Share this listing

For the vendor. Sharing assets for social media, two PNGs of 1200 × 630 that say Magic Hour MCP is listed on Anchor Terminal, with the vendor's logo and this page's address and no grade or score.

- Dark: https://www.anchorterminal.com/assets/share/magic-hour-mcp-dark.png
- Light: https://www.anchorterminal.com/assets/share/magic-hour-mcp-light.png

---

Query this catalogue directly over MCP: https://www.anchorterminal.com/mcp (Streamable HTTP, no key). How: https://www.anchorterminal.com/agents/#mcp
