Skip to content

CCAPI MCP server

ai.ccapi/mcpccapi.aistreamable-httpv1.0.1Scanned

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

11 tools: 1 do something that can't be undone, 7 change things and 3 only read.

Tools: 0 delete, 1 irreversible, 7 write, 3 read.
  • Reads private data · 1
  • Reads text others write · 7
  • Can send data out · 1
  • 1 finding

Tools listed without sign-in. Remote posture

With Anlyon

1 tool here does something that can't be undone. Govern it.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the CCAPI MCP server can do

The CCAPI MCP server is published by ccapi.ai, a domain the MCP Registry verified. It has 11 tools. It can do things that cannot be undone, such as sending, paying or publishing (chat_completion) and create or change data (7 tools: edit_image, extend_music, generate_image, generate_lyrics and 3 more), but cannot delete data. 3 tools only read.

1 tool reads private data (list_models); 7 tools bring in text other people can write (edit_image, extend_music, generate_image and 4 more); 1 tool can send data out of your system (chat_completion). On its own it has all three parts of a prompt-injection exfiltration path: private data, text others write, and a way out.

The fixed checks matched 1 finding (annotation-conflict). Each one comes with the text it matched.

Its remote endpoint lists its tools without sign-in.

Last scanned Nov 11, 2026, version 1.0.1.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

11 tools

Can't be undone 1Claude Code: ask
  • chat_completionSend a conversation to any text model available through CCAPI (Claude, GPT, Gemini, DeepSeek, GLM, MiniMax, Kimi, Qwen…) and get the reply.Can send data out1 finding

    Why it is in this group

    • The description's first verb is "send": a word for sending, and what is sent cannot be taken back.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • outboundThe description's first verb is "send": it sends or publishes something to other people.

    Findings

    • Warn

      The server marks "chat_completion" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".

      Matched text

      annotations: {"readOnlyHint":true,"idempotentHint":false}; The description's first verb is "send": a word for sending, and what is sent cannot be taken back.

      What this check looks for
    Description, as the server wrote it
    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.

    Annotations the server declares

    • idempotentHint: false
    • readOnlyHint: true
Changes things 7Claude Code: ask
  • edit_imageEdit existing images according to an instruction, using the dedicated image edit endpoint.Reads text others writeNo findings

    Why it is in this group

    • The name "edit_image" says "edit": a word for a change.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    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.

    Annotations the server declares

    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
    • readOnlyHint: false
  • extend_musicContinue an existing song from a given timestamp, producing a longer version.Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    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.

    Annotations the server declares

    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
    • readOnlyHint: false
  • generate_imageGenerate an image from a text prompt, optionally guided by reference images (image-to-image).Reads text others writeNo findings

    Why it is in this group

    • The name "generate_image" says "generate": a word for a change.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    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.

    Annotations the server declares

    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
    • readOnlyHint: false
  • generate_lyricsWrite song lyrics from a description, without generating audio.Reads text others writeNo findings

    Why it is in this group

    • The name "generate_lyrics" says "generate": a word for a change.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    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.

    Annotations the server declares

    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
    • readOnlyHint: false
  • generate_musicGenerate a song with vocals or instrumental.Reads text others writeNo findings

    Why it is in this group

    • The name "generate_music" says "generate": a word for a change.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    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.

    Annotations the server declares

    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
    • readOnlyHint: false
  • generate_videoGenerate a video from a text prompt, optionally driven by reference images (image-to-video, first/last frame).Reads text others writeNo findings

    Why it is in this group

    • The name "generate_video" says "generate": a word for a change.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    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.

    Annotations the server declares

    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
    • readOnlyHint: false
  • upscale_imageIncrease the resolution of an existing image 2x with ccapi-upscale-v1.Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    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.

    Annotations the server declares

    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
    • readOnlyHint: false
Only reads 3Show 3 read-only tools
  • get_balanceReport the remaining and used balance for this CCAPI key.No findings

    Why it is in this group

    • The name "get_balance" says "get": a word for reading.
    Description, as the server wrote it
    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.

    Annotations the server declares

    • idempotentHint: false
    • readOnlyHint: true
  • get_taskCheck the status of an asynchronous generation task submitted by generate_video, generate_music, generate_lyrics, extend_music, or generate_image with async=true.No findings

    Why it is in this group

    • The name "get_task" says "get": a word for reading.
    Description, as the server wrote it
    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.

    Annotations the server declares

    • idempotentHint: false
    • readOnlyHint: true
  • list_modelsList the AI models this CCAPI key can actually call, with their capability category and the MCP tool that drives them.Reads private dataNo findings

    Why it is in this group

    • The name "list_models" says "list": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "list" and "drive": it reads files.
    Description, as the server wrote it
    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.

    Annotations the server declares

    • idempotentHint: false
    • readOnlyHint: true

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

7 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 1 outbound tool breaks every one.

Reads private data1
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write7
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out1
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
deny to break every path
{
  "permissions": {
    "deny": [
      "mcp__ccapi__chat_completion"
    ]
  }
}

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 20, 2026, 18:25 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
8
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
not observed
Authorization servers
not observed
Legacy HTTP+SSE transport
no
Dynamic client registration
not observed
Client ID metadata documents
not observed
  • HighThe whole server

    The endpoint served tools/list without credentials, and the list has 8 tools that change things (1 irreversible, 7 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://api.ccapi.ai/mcp: tools/list returned 11 tools to an anonymous client, including chat_completion (irreversible), edit_image (write), extend_music (write), generate_image (write), generate_lyrics (write) and 3 more

    What this check looks for

Findings

Matches of fixed checks, with the text they matched.

1 warn
  • The server marks "chat_completion" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".

    Matched text

    annotations: {"readOnlyHint":true,"idempotentHint":false}; The description's first verb is "send": a word for sending, and what is sent cannot be taken back.

    What this check looks for

History

Changes to its tool definitions and to its remote endpoint, newest first.

No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.

Install

Pinned to version 1.0.1. Values in angle brackets are yours to fill in.

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

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__ccapi__get_balance",
      "mcp__ccapi__get_task",
      "mcp__ccapi__list_models"
    ],
    "ask": [
      "mcp__ccapi__chat_completion",
      "mcp__ccapi__edit_image",
      "mcp__ccapi__extend_music",
      "mcp__ccapi__generate_image",
      "mcp__ccapi__generate_lyrics",
      "mcp__ccapi__generate_music",
      "mcp__ccapi__generate_video",
      "mcp__ccapi__upscale_image"
    ],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as ccapi. Review it before you use it.

Capability tags behind 9 rules
  • mcp__ccapi__chat_completion outbound
  • mcp__ccapi__edit_image untrusted-input
  • mcp__ccapi__extend_music untrusted-input
  • mcp__ccapi__generate_image untrusted-input
  • mcp__ccapi__generate_lyrics untrusted-input
  • mcp__ccapi__generate_music untrusted-input
  • mcp__ccapi__generate_video untrusted-input
  • mcp__ccapi__list_models private-data
  • mcp__ccapi__upscale_image untrusted-input

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/ai.ccapi/mcp/verify?hash=f0101d0474f683d71d1634d5e653f91c872c27b2ed4300881ce87a66f51b7568" | jq -e '.changed == false'

Pinned hash f0101d0474f6, from Nov 11, 2026, 17:18 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/ai.ccapi/mcp/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for ai.ccapi/mcp
[![Anlyon Radar](https://radar.anlyon.com/badge/ai.ccapi/mcp.svg)](https://radar.anlyon.com/servers/ai.ccapi/mcp)

Source: MCP Registry. Synced Oct 21, 2026, 17:41 UTC.

CCAPI MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the CCAPI MCP server safe?

No scan can promise that. What its definitions show: 11 tools, 8 of which can change, send or delete something, and 1 finding from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the CCAPI MCP server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the CCAPI MCP server do things that cannot be undone?

Yes. 1 tool can send, pay, publish or merge: chat_completion.

What tools does the CCAPI MCP server have?

11 tools: 1 irreversible, 7 write, 3 read. chat_completion, edit_image, extend_music, generate_image, generate_lyrics, generate_music, generate_video, upscale_image and 3 more.

Does the CCAPI MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the CCAPI MCP server send data to outside services?

1 tool can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): chat_completion.

Has the CCAPI MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Nov 11, 2026.