apimodels
NPM · APIMODELS-MCP · 2 COMPONENTS
Image, video, chat and speech models with one account — local (npx) or hosted with OAuth sign-in.
Available components
Last scanned (3 hours ago)
Scores higher than 55% of npm servers
Category AI and Machine Learning
How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. How we score → Why this is hard to score →
Supply Chain Security81
- No malware found by supply-chain analysis.Pass
- CVE check failed: a known high-severity CVE affects sharp 0.34.5, a direct dependency. A fixed version is available. View diagnostics → Fail
- No install/post-install scripts declared.Pass
- 30 of 123 dependencies flagged as unhealthy. View diagnostics → Partial
Provenance & Transparency45
- Source repository is publicly reachable at the declared URL. View diagnostics → Pass
- Provenance check failed: no build-provenance attestation is published. See how to fix → View diagnostics → Fail
- Clear OSI-approved license (MIT).Pass
- Actively maintained (last published 25 days ago).Pass
- Disclosure check failed: no security disclosure policy was found in the source repository. See how to fix → Fail
Schema Quality & AI Usability63
- AI-judged instruction clarity (excellent).Pass
- Context-footprint check failed: tool/resource definitions use about 2327 tokens (~290/item across 8 items; 8 tools + 0 resources), over budget; trim descriptions and params. See how to fix → Fail
- Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management100
- No destabilizing schema changes in the last 30 days.Pass
Tool Coverage100
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 100% of tool parameters carry a description.Pass
Tool Safety100
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- We read all 8 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
- An AI judge read all 8 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
- Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.
- 29 Oct 26 +1
- Stability: 0.97 → pass security
- 28 Oct 26 −1
- Stability: pass → 0.97 functional
- 24 Oct 26 0
- CVE-2023-54404 affects this package: high ▼ security
- 22 Oct 26 0
- CVE-2026-114113 affects this package: high ▼ security
- CVE-2026-114111 affects this package: high ▼ security
- CVE-2026-114112 affects this package: high ▼ security
- GHSA-f88m-g3jw-g9cj no longer affects this package ▲ security
- GHSA-rgj7-g3m4-5g8c no longer affects this package ▲ security
- GHSA-wq5f-xc86-pv6w no longer affects this package ▲ security
- 18 Oct 26 +1
- Stability: 0.97 → pass security
- 16 Oct 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 90 to 93. That category is still filling its 30-day observation window: 27 days of observed history at the previous scan, 28 at this one. The score rises as the window fills, whether or not the server changes.
- 14 Oct 26 +42
- Injection markers: unverified → pass ▲ security
- Tool coverage: unverified → 100 ▲ functional
- MCP protocol: unverified → pass ▲ functional
- Stability: unverified → 0.87 ▲ functional
- Package version: 0.4.1 → 0.7.0 functional
- 13 Oct 26 0
- Security disclosure: unverified → fail ▼ functional
- Package version: 0.4.1 → 0.7.0 functional
Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.
Captured 5 Nov 2026 · Analysed npm/apimodels-mcp@0.7.0
Provenance No attestation
The registry publishes no build provenance for this version, so there is nothing to verify.
| Result | No attestation |
|---|---|
| Ecosystem | npm |
Background: How many MCP packages publish verified provenance →
Vulnerabilities 4 findings
| ID | CVE | Severity | Vector | Fix available |
|---|---|---|---|---|
| GHSA-f88m-g3jw-g9cj | CVE-2026-114113 | high | yes | |
| GHSA-rgj7-g3m4-5g8c | CVE-2026-114112 | high | yes | |
| GHSA-wq5f-xc86-pv6w | CVE-2026-114111 | high | yes | |
| GHSA-mrhx-c7xc-m3w5 | CVE-2023-54404 | high | CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H | yes |
Background: What a vulnerability scan can and cannot prove →
Dependencies 123 packages
| Packages resolved | 123 |
|---|---|
| Stale | 30 |
| Tree resolution | Complete |
Background: SBOMs and build attestations, explained →
The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →
chat ~192
Chat / text completion with any LLM on apimodels.app (GPT-5.5, Claude, Gemini, GLM, DeepSeek, Qwen, …). Returns the assistant reply text.
| Name | Type | Req | Description |
|---|---|---|---|
| max_tokens | integer | – | Optional max output tokens. |
| model | string | – | Model id, e.g. gpt-5-5, gpt-6.1-sol, gpt-6-sol, claude-opus-5-5, claude-sonnet-5-5 (Claude Sonnet 5.5, fast and strong at coding and agent work, $1.20 / $6 per 1M tokens), claude-sonnet-5, gemini-3-p… |
| prompt | string | yes | The user message / prompt. |
| system | string | – | Optional system prompt. |
No output schema declared.
No examples provided.
generate_image ~785
Generate an image from a text prompt (or edit an input image). Returns the URL(s) of the generated image, valid 7 days, plus a downscaled preview of the image itself when the client can show tool-result images to you. Images take about 50 s (9 in 10 within 90 s); if the task is still running when wait_seconds is up you get a task id — call get_task with it, do not resubmit. If you cannot see the image in the result, call review_image with the returned URL to get a written critique and a revised prompt, then generate again. Roughly $0.025 per image on the default model; gpt-image-2-lite is $0.008.
| Name | Type | Req | Description |
|---|---|---|---|
| aspect_ratio | string | – | Optional aspect ratio, e.g. 1:1, 16:9, 9:16. |
| background | string | – | Set to "transparent" to edit an image that already has a transparent background and keep it transparent (PNG with alpha out). Needs image_url pointing at ONE PNG with an alpha channel. doubao-seedrea… |
| image_url | string | – | Optional input image for image-to-image edits. Accepts a public https:// URL, a LOCAL FILE PATH, a localhost URL, or a data: URI — local sources are uploaded for you automatically. |
| layer_decomposition | boolean | – | Split the image in image_url into a flattened base plus up to 16 transparent PNG layers (text blocks, subjects, props), each returned with a name and its box in the original. prompt is optional. Bill… |
| model | string | – | Image model id, e.g. gpt-image-2, gpt-image-2.5-flare, flux-2-klein-4b ($0.006, fastest, up to 3 reference images), gpt-image-2-lite, gemini-3-pro-image, gemini-3-pro-image-gemini ($0.03 flat), qwen3… |
| prompt | string | – | Text description of the image to generate (or the edit to make). Required, except with layer_decomposition where it is optional. |
| resolution | string | – | Optional resolution, e.g. 1K, 2K, 4K. |
| return_image | boolean | – | Attach a downscaled preview (max 1024px JPEG) of the result so you can look at it. Set false to save context when you only need the URL. |
| wait_seconds | integer | – | How long to wait for the image before returning a task id instead (seconds, 0–900). Default 50: Codex aborts tool calls at 60 s unless tool_timeout_sec is raised. In Claude Code / Claude Desktop you… |
No output schema declared.
No examples provided.
generate_video ~478
Generate a video from a text prompt (and optional reference image). Submits the job and waits up to wait_seconds: returns the video URL(s) (valid 7 days) if it finishes in time, otherwise a task id — then call get_task with that id; never resubmit a running task. Video takes a median 2.5 minutes and 9 in 10 finish within 8 minutes, so expect one or two get_task calls. Video is the most expensive modality here — the default model costs roughly $0.30-$0.50 per clip; pass model:"veo-3.1-fast-fhd" for the cheapest option at $0.07 flat.
| Name | Type | Req | Description |
|---|---|---|---|
| aspect_ratio | string | – | Optional aspect ratio, e.g. 16:9, 9:16, 1:1. |
| duration | number|string | – | Optional duration in seconds, e.g. 5 or 10. |
| image_url | string | – | Optional first-frame / reference image for image-to-video. Accepts a public https:// URL, a LOCAL FILE PATH, a localhost URL, or a data: URI — local sources are uploaded for you automatically. |
| model | string | – | Video model id. Use the dotted public names: seedance-2.0-fast, seedance-2.0, seedance-2.0-mini, seedance-2.5, wan-3.0-video, minimax-h3, minimax-h3-lite (low cost, up to 768p), ltx-2.3, grok-imagine… |
| prompt | string | yes | Text description of the video. |
| resolution | string | – | Optional resolution, e.g. 480p, 720p, 1080p. |
| wait_seconds | integer | – | How long to wait for the video before returning a task id instead (seconds, 0–900). Default 50: Codex aborts tool calls at 60 s unless tool_timeout_sec is raised. In Claude Code / Claude Desktop you… |
No output schema declared.
No examples provided.
get_balance ~50
Show the apimodels account balance (USD) for the configured API key. Check it before submitting a batch of generations, and tell the user if the balance looks too low for what they asked for.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
get_task ~247
Check on, or wait for, an image / video / speech task that generate_image, generate_video or text_to_speech handed back as "still running". Returns the result URL(s) once it has finished (with an image preview for image tasks), a failure message if it failed, or "still running" again — in which case call get_task once more. Waits up to wait_seconds before answering, so one call usually suffices for a task that is nearly done. Task ids also appear at https://apimodels.app/console/records.
| Name | Type | Req | Description |
|---|---|---|---|
| return_image | boolean | – | For image tasks, attach a downscaled preview of the result. Set false to save context. |
| task_id | string | yes | The task id from the "STILL RUNNING" message (or from the apimodels console). |
| wait_seconds | integer | – | How long to wait for the task before returning a task id instead (seconds, 0–900). Default 50: Codex aborts tool calls at 60 s unless tool_timeout_sec is raised. In Claude Code / Claude Desktop you c… |
No output schema declared.
No examples provided.
list_models ~73
List the model ids available on apimodels.app, grouped by type (chat, image, video, audio, embedding) with the endpoint each one uses. Every id listed is callable. Use chat ids with chat, image ids with generate_image, video ids with generate_video, audio ids with text_to_speech.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
review_image ~217
Have a vision model look at an image and critique it against a brief. Returns what matches, what is wrong (garbled text, composition, colors, artifacts) and a revised prompt. Use it after generate_image to check the result and decide whether to regenerate — this works in every MCP client, including ones that do not pass tool-result images to you. Costs one small vision chat call (well under $0.01 on the default model).
| Name | Type | Req | Description |
|---|---|---|---|
| brief | string | yes | What the image is supposed to show — usually the prompt it was generated from, plus any requirements the user stated (exact text, aspect ratio, style). |
| image_url | string | yes | The image to review: a URL returned by generate_image, any public https:// URL, a LOCAL FILE PATH, a localhost URL, or a data: URI. |
| model | string | – | Vision-capable chat model that does the looking. gpt-5.6-luna (default, cheapest) or claude-sonnet-5 for a more careful read. |
No output schema declared.
No examples provided.
text_to_speech ~285
Convert text to speech (MiniMax voices). Returns the audio file URL (valid 7 days). Costs about $0.004 for a short line; billed at $0.04 per 1000 characters.
| Name | Type | Req | Description |
|---|---|---|---|
| model | string | – | TTS model id, e.g. minimax-speech-02-turbo (fast), minimax-speech-02-hd / minimax-speech-2.8-hd (higher quality). Note: eleven-tts-* models are NOT available here — they stream from POST /v1/tts/stre… |
| speed | number | – | Optional speaking rate, 0.5-2 (1 = normal). |
| text | string | yes | The text to speak. |
| voice_id | string | – | Voice id. English: English_Trustworthy_Man, English_Graceful_Lady, Serene_Woman. Chinese: male-qn-qingse, female-tianmei. Full list: GET /v1/minimax/voices. |
| wait_seconds | integer | – | How long to wait for the audio before returning a task id instead (seconds, 0–900). Default 50: Codex aborts tool calls at 60 s unless tool_timeout_sec is raised. In Claude Code / Claude Desktop you… |
No output schema declared.
No examples provided.
How do I install the apimodels MCP server?
apimodels runs locally as an npm package, launched with npx -y apimodels-mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.
npm · apimodels-mcp
claude mcp add stimqq-apimodels-mcp -- npx -y apimodels-mcp
What is the apimodels MCP server?
apimodels is an MCP server listed in the public MCP registry as io.github.stimQQ/apimodels-mcp. Image, video, chat and speech models with one account, local (npx) or hosted with OAuth sign-in. This page covers its npm package (apimodels-mcp).
Is the apimodels MCP server safe to use?
apimodels scores 77 out of 100 on VerifyMCP. We recorded 4 known advisories against it as of 5 November 2026. It declares no install or post-install scripts. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.
What tools does the apimodels MCP server expose?
apimodels exposes 8 tools: get_balance, list_models, chat, generate_image, review_image, and 3 more. Their descriptions and schemas cost roughly 2,327 tokens of context every time the server is loaded.
Is the apimodels MCP server still maintained?
apimodels is still listed as active in the MCP registry. We last reached this channel on 5 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.
What licence is the apimodels MCP server under?
apimodels declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.