# CCAPI (remote · api.ccapi.ai)

Image, video, music and text generation across 100+ models through one endpoint.

- Trust score: 73/100 (medium)
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-08-03

## Components

- remote · `api.ccapi.ai`: 73/100 (this document), [markdown](https://verifymcp.io/servers/ai-ccapi-mcp/api.md), [page](https://verifymcp.io/servers/ai-ccapi-mcp/api)

## Channel facts

- Endpoint: `https://api.ccapi.ai/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.0.1`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-08-03.

- **Endpoint Security**: 89/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation is enforced on tool calls, but the challenge carries no valid RFC 9728 metadata, so a client cannot discover where to get a token.
  - HTTPS is enforced; there's no plaintext access path.
  - The HSTS (Strict-Transport-Security) header is present.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 61/100
  - AI-judged instruction clarity (excellent).
  - Context-footprint check failed: tool/resource definitions use about 2785 tokens (~253/item across 11 items; 11 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 3/100
  - Stability observed for 1 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
  - Structured output schemas are declared (100% of tools); any adoption earns full credit.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### Claude

```bash
claude mcp add --transport http ai-ccapi-mcp https://api.ccapi.ai/mcp
```

### Codex

```toml
[mcp_servers.ai-ccapi-mcp]
url = "https://api.ccapi.ai/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "ai-ccapi-mcp": {
      "type": "remote",
      "url": "https://api.ccapi.ai/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add ai-ccapi-mcp --url https://api.ccapi.ai/mcp --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  ai-ccapi-mcp:
    url: "https://api.ccapi.ai/mcp"
```

### Other

```json
{
  "mcpServers": {
    "ai-ccapi-mcp": {
      "type": "http",
      "url": "https://api.ccapi.ai/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-08-03 (score 73, 0)

- [functional improvement] Stability: unverified → 0.03

### 2026-08-02 (score 73)

First indexed and scored.

## MCP tools (11)

### `chat_completion` (~231 tokens)

Call a text model

Send a conversation to any text model available through CCAPI (Claude, GPT, Gemini, DeepSeek, GLM, MiniMax, Kimi, Qwen…) and get the reply. Useful for consulting a second model for a different perspective, running a cheap model over bulk work, or reaching a model the current client does not have configured. Responses are non-streaming.

Input parameters:

- `max_tokens` (null|integer): Maximum tokens to generate. Omit for the model default.
- `messages` (null|array, required): Conversation so far, oldest first. Must contain at least one message.
- `model` (string, required): Exact model name, e.g. claude-sonnet-5 or deepseek-v4-pro. Call list_models with category 'text' if unsure.
- `stop` (null|array): Up to 4 sequences that stop generation.
- `temperature` (null|number): Sampling temperature, typically 0-2. Omit for the model default.
- `top_p` (null|number): Nucleus sampling probability mass, 0-1. Omit for the model default.

Output parameters:

- `content` (string): The model's reply text.
- `finish_reason` (string): Why generation stopped, e.g. stop or length.
- `model` (string): Model that actually served the request.
- `usage` (object): Token accounting for this call.

### `edit_image` (~398 tokens)

Edit an image

Edit existing images according to an instruction, using the dedicated image edit endpoint. Provide at least one source image. For pure text-to-image, or for models that take references through the generation endpoint, use generate_image instead.

Input parameters:

- `aspect_ratio` (string): Z-Image only: required output aspect ratio. Supported values: 1:1, 4:3, 3:4, 16:9, 9:16.
- `async` (boolean): Submit asynchronously and return a task_id instead of waiting. Use when the client times out on slow models; poll with get_task using platform 'image'.
- `images` (null|array): Optional reference images as HTTPS URLs or data URLs, for image-to-image generation. Not all models accept references.
- `model` (string, required): Exact model name, e.g. gpt-image-2 or nano-banana-pro. Call list_models with category 'image' if unsure.
- `n` (null|integer): How many images to generate. Defaults to 1. Each one is billed.
- `nsfw_checker` (null|boolean): Z-Image only: enable or disable the upstream NSFW checker. Explicit false is preserved.
- `prompt` (string, required): Description of the image to generate.
- `quality` (string): Quality hint, e.g. low, medium, high or auto. Model-dependent.
- `response_format` (string): 'url' (default) returns links; 'b64_json' returns base64 data, which is far larger and usually unnecessary.
- `size` (string): Output size, e.g. 1024x1024, 1536x1024, or 'auto'. Model-dependent; omit for the model default.
- `wait_seconds` (null|integer): Only with async: poll server-side for up to this many seconds (max 240) and return the finished image if it completes in time.

Output parameters:

- `images` (null|array): Generated images, when the call completed.
- `model` (string): Model that served the request.
- `note` (string): What to do next.
- `platform` (string): Pass this to get_task together with task_id.
- `status` (string): One of: completed, queued, in_progress, failed.
- `task_id` (string): Task identifier, when submitted asynchronously and still running.

### `extend_music` (~182 tokens)

Extend a song

Continue an existing song from a given timestamp, producing a longer version. Requires the clip_id of a previously generated song, which appears in the 'raw' payload of a completed generate_music task. Asynchronous — poll with get_task.

Input parameters:

- `clip_id` (string, required): ID of the song to extend, taken from the 'raw' clip list of a completed generate_music task.
- `continue_at` (number, required): Timestamp in seconds within the original song to continue from.
- `prompt` (string): Lyrics for the extended section.
- `provider` (string): Music backend: 'suno' (default) or 'producer'.
- `tags` (string): Style tags for the extension.
- `title` (string): New title for the extended song.
- `wait_seconds` (null|integer): Poll server-side for up to this many seconds (max 240).

Output parameters:

- `audio_urls` (null|array): Finished audio URLs, when the task completed within wait_seconds.
- `note` (string): What to do next.
- `platform` (string): 'suno' or 'producer'. Pass to get_task.
- `raw`: Full result payload when completed — clip IDs, lyrics text and per-clip metadata. Clip IDs from here are what extend_music takes.
- `status` (string): One of: queued, in_progress, completed, failed.
- `task_id` (string): Pass this to get_task together with platform to check progress.

### `generate_image` (~418 tokens)

Generate an image

Generate an image from a text prompt, optionally guided by reference images (image-to-image). Returns image URLs directly. Most models finish in 10-60 seconds; if your client times out, set async=true to get a task_id you can poll instead. This call costs money — do not retry speculatively.

Input parameters:

- `aspect_ratio` (string): Z-Image only: required output aspect ratio. Supported values: 1:1, 4:3, 3:4, 16:9, 9:16.
- `async` (boolean): Submit asynchronously and return a task_id instead of waiting. Use when the client times out on slow models; poll with get_task using platform 'image'.
- `images` (null|array): Optional reference images as HTTPS URLs or data URLs, for image-to-image generation. Not all models accept references.
- `model` (string, required): Exact model name, e.g. gpt-image-2 or nano-banana-pro. Call list_models with category 'image' if unsure.
- `n` (null|integer): How many images to generate. Defaults to 1. Each one is billed.
- `nsfw_checker` (null|boolean): Z-Image only: enable or disable the upstream NSFW checker. Explicit false is preserved.
- `prompt` (string, required): Description of the image to generate.
- `quality` (string): Quality hint, e.g. low, medium, high or auto. Model-dependent.
- `response_format` (string): 'url' (default) returns links; 'b64_json' returns base64 data, which is far larger and usually unnecessary.
- `size` (string): Output size, e.g. 1024x1024, 1536x1024, or 'auto'. Model-dependent; omit for the model default.
- `wait_seconds` (null|integer): Only with async: poll server-side for up to this many seconds (max 240) and return the finished image if it completes in time.

Output parameters:

- `images` (null|array): Generated images, when the call completed.
- `model` (string): Model that served the request.
- `note` (string): What to do next.
- `platform` (string): Pass this to get_task together with task_id.
- `status` (string): One of: completed, queued, in_progress, failed.
- `task_id` (string): Task identifier, when submitted asynchronously and still running.

### `generate_lyrics` (~147 tokens)

Generate lyrics

Write song lyrics from a description, without generating audio. Useful as a first step before generate_music with custom=true, so the user can approve the words before paying for audio. Asynchronous — poll with get_task.

Input parameters:

- `lyrics_model` (string): Lyrics model: classic or remi.
- `prompt` (string, required): Description of the lyrics to write, e.g. 'a love song about meeting someone at a coffee shop'.
- `provider` (string): Music backend: 'suno' (default) or 'producer'.
- `wait_seconds` (null|integer): Poll server-side for up to this many seconds (max 240). Lyrics usually finish in under 30 seconds.

Output parameters:

- `audio_urls` (null|array): Finished audio URLs, when the task completed within wait_seconds.
- `note` (string): What to do next.
- `platform` (string): 'suno' or 'producer'. Pass to get_task.
- `raw`: Full result payload when completed — clip IDs, lyrics text and per-clip metadata. Clip IDs from here are what extend_music takes.
- `status` (string): One of: queued, in_progress, completed, failed.
- `task_id` (string): Pass this to get_task together with platform to check progress.

### `generate_music` (~338 tokens)

Generate a song

Generate a song with vocals or instrumental. Two modes: simple (custom=false) where you describe the song and the model writes the lyrics, or custom (custom=true) where you supply lyrics, title and style tags. Asynchronous — returns a task_id you poll with get_task; generation usually takes 30-120 seconds and normally yields two variations.

Input parameters:

- `custom` (boolean): false (default) lets the model write the lyrics from your description; true means prompt contains lyrics you wrote and you should also set title and tags.
- `instrumental` (boolean): Generate without vocals.
- `model` (string): Suno model version, e.g. chirp-v5 or chirp-v4-5-plus. Omit for the platform default.
- `negative_tags` (string): Styles to avoid, e.g. 'heavy metal, rap'.
- `prompt` (string, required): In simple mode (custom=false) a description of the song, e.g. 'a dreamy lo-fi track about late-night trains'. In custom mode (custom=true) the actual lyrics, with section markers like [Verse] and [Ch…
- `provider` (string): Music backend: 'suno' (default) or 'producer'.
- `tags` (string): Comma-separated style tags, e.g. 'pop, upbeat, guitar'. Max 200 characters.
- `title` (string): Song title. Used in custom mode.
- `wait_seconds` (null|integer): Poll server-side for up to this many seconds (max 240) and return finished audio if it completes in time. Omit to return a task_id immediately.

Output parameters:

- `audio_urls` (null|array): Finished audio URLs, when the task completed within wait_seconds.
- `note` (string): What to do next.
- `platform` (string): 'suno' or 'producer'. Pass to get_task.
- `raw`: Full result payload when completed — clip IDs, lyrics text and per-clip metadata. Clip IDs from here are what extend_music takes.
- `status` (string): One of: queued, in_progress, completed, failed.
- `task_id` (string): Pass this to get_task together with platform to check progress.

### `generate_video` (~398 tokens)

Generate a video

Generate a video from a text prompt, optionally driven by reference images (image-to-video, first/last frame). This is asynchronous: it returns a task_id you poll with get_task(platform='video'). Generation usually takes 30-180 seconds. Pass wait_seconds to have the server poll for you. Video generation is the most expensive capability here — confirm the prompt with the user before spending on retries.

Input parameters:

- `duration` (null|integer): Clip length in seconds. Model-dependent, typically 4-15; an unsupported value is rejected with a message naming the supported ones. Omit for the model default.
- `generate_audio` (null|boolean): Generate a synchronised soundtrack. Only some models support this.
- `image_roles` (null|array): Role of each entry in images, positionally: first_frame, last_frame or reference_image.
- `images` (null|array): Optional reference image URLs. A single image is used as the first frame; use image_roles to control multiple.
- `model` (string, required): Exact model name, e.g. doubao-seedance-2-0-260128, veo-3.1-quality or viduq3-pro. Call list_models with category 'video' if unsure.
- `prompt` (string, required): Description of the video to generate.
- `ratio` (string): Aspect ratio, e.g. 16:9, 9:16, 1:1, 21:9 or adaptive. Which values a model accepts varies; an unsupported one is rejected with a message naming the valid ones.
- `resolution` (string): Output resolution: 480p, 720p (widest compatibility) or 1080p. Model-dependent.
- `wait_seconds` (null|integer): Poll server-side for up to this many seconds (max 240) and return the finished video if it completes in time. Omit to return a task_id immediately.

Output parameters:

- `model` (string): Model that accepted the request.
- `note` (string): What to do next.
- `platform` (string): Always 'video'. Pass to get_task.
- `status` (string): One of: queued, in_progress, completed, failed.
- `task_id` (string): Pass this to get_task together with platform to check progress.
- `video_urls` (null|array): Finished video URLs, when the task completed within wait_seconds.

### `get_balance` (~39 tokens)

Get account balance

Report the remaining and used balance for this CCAPI key. Useful before starting an expensive batch of generations, or to explain a quota failure to the user.

Output parameters:

- `currency` (string): Unit of the amounts above, e.g. USD.
- `note` (string): Guidance when the balance is low, unlimited, or scoped to a token.
- `remaining_amount` (number): Remaining balance, in the account's display currency. Omitted when unlimited.
- `total_amount` (number): Remaining plus used. Omitted when unlimited.
- `unlimited` (boolean): True when this token has no spending cap, in which case the amount fields are not meaningful.
- `used_amount` (number): Balance consumed to date, in the same unit.

### `get_task` (~166 tokens)

Check an async task

Check the status of an asynchronous generation task submitted by generate_video, generate_music, generate_lyrics, extend_music, or generate_image with async=true. Pass the platform and task_id exactly as they were returned. Status is one of queued, in_progress, completed or failed; poll every few seconds until it settles, or pass wait_seconds to have the server wait for you.

Input parameters:

- `platform` (string, required): Platform the task belongs to, exactly as returned by the submit tool: image, video, suno or producer.
- `task_id` (string, required): Task identifier returned by the submit tool.
- `wait_seconds` (null|integer): Poll server-side for up to this many seconds (max 240) and return once the task finishes. Omit for a single immediate status check.

Output parameters:

- `fail_reason` (string): Why the task failed, when status is failed.
- `note` (string): What to do next.
- `platform` (string): Platform the task belongs to.
- `progress` (string): Progress indication, when the backend reports one.
- `raw`: Backend payload, for details this summary does not cover (lyrics text, per-clip metadata, seeds).
- `result_urls` (null|array): Finished asset URLs. May expire; download anything worth keeping.
- `status` (string): One of: queued, in_progress, completed, failed, unknown.
- `task_id` (string): The task this status refers to.

### `list_models` (~128 tokens)

List available models

List the AI models this CCAPI key can actually call, with their capability category and the MCP tool that drives them. Call this before generating anything if you are unsure a model name is valid — availability depends on the key's group and changes over time. Never guess model names.

Input parameters:

- `category` (string): Filter by capability. One of: text, image, video, music, embedding, other. Omit to list every model this key can use.
- `search` (string): Case-insensitive substring match on the model name, e.g. 'claude' or 'seedance'.

Output parameters:

- `models` (null|array): Models available to the calling key.
- `note` (string): Guidance about the result set.
- `total` (integer): Number of models returned.

### `upscale_image` (~145 tokens)

Upscale an image

Increase the resolution of an existing image 2x with ccapi-upscale-v1. Input must be between 256x256 and 1024x1024 and under 5 MB; for larger inputs, regenerate with a high-resolution model instead.

Input parameters:

- `image` (string, required): The image to upscale, as an HTTPS URL, data URL, or raw base64. Width and height must each be 256-1024 px and the file at most 5 MB.
- `model_quality` (string): Upscale quality hint: HQ, MQ (default) or LQ.
- `response_format` (string): 'url' (default) or 'b64_json'.

Output parameters:

- `images` (null|array): Generated images, when the call completed.
- `model` (string): Model that served the request.
- `note` (string): What to do next.
- `platform` (string): Pass this to get_task together with task_id.
- `status` (string): One of: completed, queued, in_progress, failed.
- `task_id` (string): Task identifier, when submitted asynchronously and still running.

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/ai-ccapi-mcp/api#diagnostics

## Score history

- 2026-08-03: 73
- 2026-08-02: 73

## Links

- Remote endpoint: https://api.ccapi.ai/mcp
- Website: https://ccapi.ai/mcp
- Changelog RSS feed: https://verifymcp.io/servers/ai-ccapi-mcp/api/changelog.xml
- Changelog JSON feed: https://verifymcp.io/servers/ai-ccapi-mcp/api/changelog.json
- HTML version of this page: https://verifymcp.io/servers/ai-ccapi-mcp/api