# com.zipirapp/zipir (remote · api.zipirapp.com)

Turn documents into source-cited video lessons for safety, procedure and training teams.

- Trust score: 80/100 (high trust)
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- remote · `api.zipirapp.com`: 80/100 (this document), [markdown](https://verifymcp.io/servers/com-zipirapp-zipir/api.md), [page](https://verifymcp.io/servers/com-zipirapp-zipir/api)
- npm · `@zipir/mcp`: 64/100, [markdown](https://verifymcp.io/servers/com-zipirapp-zipir/zipir-mcp.md), [page](https://verifymcp.io/servers/com-zipirapp-zipir/zipir-mcp)

## Channel facts

- Endpoint: `https://api.zipirapp.com/mcp`
- Transports: `streamable-http`
- Auth: `required`
- Version: `0.1.5`

## 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 2026-11-05.

- **Endpoint Security**: 94/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - Authorisation is enforced on tool calls, advertised via RFC 9728 protected-resource metadata. Discovery is public, which costs nothing: no tool can be invoked without 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.
  - The authorisation server offers only Dynamic Client Registration (RFC 7591), which MCP 2026-07-28 deprecated in favour of Client ID Metadata Documents.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 72/100
  - AI-judged instruction clarity (excellent).
  - Context-footprint check failed: tool/resource definitions use about 1422 tokens (~177/item across 8 items; 8 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 10/100
  - Stability observed for 3 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.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 8 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 9 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the com.zipirapp/zipir MCP server?

com.zipirapp/zipir is a hosted endpoint at https://api.zipirapp.com/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http com-zipirapp-zipir 'https://api.zipirapp.com/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "com-zipirapp-zipir": {
      "url": "https://api.zipirapp.com/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "com-zipirapp-zipir": {
      "type": "http",
      "url": "https://api.zipirapp.com/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.com-zipirapp-zipir]
url = "https://api.zipirapp.com/mcp"
```

### opencode

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

### OpenClaw

```bash
openclaw mcp add com-zipirapp-zipir --url 'https://api.zipirapp.com/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  com-zipirapp-zipir:
    url: "https://api.zipirapp.com/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "com-zipirapp-zipir": {
      "Transport": "http",
      "Url": "https://api.zipirapp.com/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add com-zipirapp-zipir -t streamable-http -u 'https://api.zipirapp.com/mcp'
```

### Other

```json
{
  "mcpServers": {
    "com-zipirapp-zipir": {
      "type": "http",
      "url": "https://api.zipirapp.com/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.

### 2026-11-04 (score 80, 0)

- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “get_usage” rewrote its description, which is the text the model reads
- [functional regression] Schema quality: 148 → 177
- [functional regression] Schema quality: 148 → 175
- [functional] Server version: 1.0.0 → 0.1.5
- [functional] New tool “list_voices”
- [cosmetic] “create_lesson” added an optional parameter “lang”
- [cosmetic] “create_lesson” added an optional parameter “voice”
- [cosmetic] Tool “list_voices” changed its title: List narrator voices

### 2026-11-03 (score 80, +1)

- [security] Tool “create_lesson” rewrote its description, which is the text the model reads
- [functional improvement] Stability: unverified → 0.03

### 2026-11-02 (score 79, +41)

- [security improvement] Transport: unverified → pass
- [security improvement] Injection markers: unverified → pass
- [security] First check of Judged manipulation: pass
- [security] Authorization: Authorisation is enforced on tool calls, advertised via RFC 9728 protected-resource metadata. Discovery is public, which costs nothing: no tool can be invoked without a token.
- [functional improvement] MCP protocol: unverified → pass
- [functional improvement] Tool coverage: unverified → 100
- [functional] First check of Destructive annotations: pass
- [functional] First check of Schema quality: fail
- [functional] First check of Schema quality: excellent
- [functional] First check of Schema quality: fail
- [functional] First check of Tool coverage: 100

### 2026-10-31 (score 38)

First indexed and scored.

## MCP tools (8)

### `list_formats` (~78 tokens)

List video formats

List video formats (call this before create_lesson when the user names a style)

Use when someone asks which styles exist, or before create_lesson if they said a style ("short", "explainer", "vertical"). Returns format ids to pass to create_lesson; never hard-code an id, read it here. No minutes are spent.

### `list_voices` (~136 tokens)

List narrator voices

List narrator voices (call this before create_lesson when the user names a voice, a tone or a gender)

Use when someone asks which voices exist, wants a female or male narrator, a calm or lively tone, or English narration. Returns each voice with its name, a one-sentence tone, language (tr or en), gender, accent, tier and a short audio sample. Pass a voice `key` to create_lesson. Optional `language` narrows the list. Premium voices spend 1.5x minutes. No minutes are spent here.

Input parameters:

- `language` (string): Only voices that speak this language.

### `list_lessons` (~147 tokens)

List my lessons

List this account's recent lessons (use when someone asks "my videos", "what did I make", or wants to continue an earlier one)

Use when the user refers to their own earlier work: their videos, the lesson they started yesterday, the one waiting for approval. Returns the newest lessons with id, title, status, minutes and a link to open each. Read-only and free. Pass `status` to narrow (queued | rendering | done | failed) and `limit` (default 20, max 50).

Input parameters:

- `limit` (integer): How many to return (default 20, max 50).
- `status` (string): Only lessons in this state.

### `create_lesson` (~473 tokens)

Create a lesson

Create a narrated video lesson from a document, text or web page (use when someone wants a document turned into a training, safety or procedure video)

Use when the user brings a document, a block of text or a web page and wants a video, a lesson, an explainer or a summary video of it. Accepted sources: a file (PDF, DOCX, PPTX, MD or TXT, max 50 MB) sent as multipart, which the local MCP server takes as `file_path`; plain `text` (max 5 MB); or a public http(s) `url` whose readable text becomes the document. Every on-screen sentence cites the source paragraph it came from, so use this when the user needs traceable training, safety or procedure videos. Narration is Turkish. Returns a lesson id immediately; nothing renders and no minutes are spent until approve_lesson. Follow it with get_lesson until the storyboard needs approval. With an Idempotency-Key the same request inside 24h returns the same lesson (same key, different body is 409). API documentation: https://zipirapp.com/en/api/

Input parameters:

- `format` (string): Video format. GET /v1/formats lists the live set.
- `lang` (string): Language of the video's narration and on-screen text. Omit it and an English document gets an English video, any other document a Turkish one. A document in another language than the one you ask for…
- `text` (string): The lesson text itself (UTF-8, max 5 MB).
- `title` (string): Shown in the panel and used in the download file name.
- `url` (string): A public http(s) page; its readable text becomes the document.
- `voice` (string): Narrator voice key from GET /v1/voices. A voice that declares a language (the premium ones) also sets `lang`; a `lang` that differs is a 400 voice_lang_mismatch. The standard voices declare none and…

### `get_lesson` (~110 tokens)

Get lesson status

Check a lesson's status (poll this after create_lesson and after approve_lesson)

Use whenever you need to know where a lesson stands: after creating it, while it renders, or before downloading. `status` is the machine state (queued | rendering | done | failed), `phase` is finer, `needs_approval` is true while the storyboard waits for a human, and `next` says which call to make next. Costs no minutes.

Input parameters:

- `id` (string, required): id

### `approve_lesson` (~95 tokens)

Approve and render (spends minutes)

Approve the storyboard and start rendering (the only call that spends the account's minutes)

Use only after the user has seen the storyboard and agreed: this is the panel's approval button. The minutes for the lesson are held at this call and the render is queued. Do not approve on the user's behalf without their go-ahead. Check get_usage first if the balance may be tight.

Input parameters:

- `id` (string, required): id

### `get_download` (~115 tokens)

Get download link

Get a signed download link for the finished video (use when the video is ready and the user wants the file)

Use once get_lesson says the video is done. Returns a URL (valid 900 s) to the MP4, not the file itself: hand the link to the user or fetch it. `variant=720` is the light copy every plan gets, `variant=1080` is full quality and needs a paid plan.

Input parameters:

- `id` (string, required): id
- `variant` (string): variant

### `get_usage` (~150 tokens)

Get minute balance

Check remaining video minutes before spending any (call this before approve_lesson if unsure)

Use when the user asks how much is left, or before approve_lesson when the balance may be tight. `remaining` is what this key can spend now: the account's own balance as the billing page shows it (plan minutes left plus bought packs, or the free trial plus packs), narrowed by the key's own monthly cap if it has one; never a second rule. `used` is this key's minutes in the current period (finished videos plus minutes held for its lessons still rendering); `limit` = used + remaining. approve_lesson answers 402 quota_exceeded when the lesson does not fit.

## Diagnostics

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

## Score history

- 2026-11-05: 80
- 2026-11-04: 80
- 2026-11-03: 80
- 2026-11-02: 79
- 2026-11-01: 38
- 2026-10-31: 38

## Common questions

### What is the com.zipirapp/zipir MCP server?

com.zipirapp/zipir is an MCP server listed in the public MCP registry as com.zipirapp/zipir. Turn documents into source-cited video lessons for safety, procedure and training teams. This page covers its hosted endpoint (https://api.zipirapp.com/mcp).

### Is the com.zipirapp/zipir MCP server safe to use?

com.zipirapp/zipir scores 80 out of 100 on VerifyMCP. 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 com.zipirapp/zipir MCP server expose?

com.zipirapp/zipir exposes 8 tools: list_formats, list_voices, list_lessons, create_lesson, get_lesson, and 3 more. Their descriptions and schemas cost roughly 1,304 tokens of context every time the server is loaded.

### Does the com.zipirapp/zipir MCP server require authentication?

Yes. com.zipirapp/zipir asked us for credentials when we connected, so you will need to authorise it in your MCP client before it can do anything.

### Is the com.zipirapp/zipir MCP server still maintained?

com.zipirapp/zipir 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.

## Links

- Remote endpoint: https://api.zipirapp.com/mcp
- Changelog RSS feed: https://verifymcp.io/servers/com-zipirapp-zipir/api.xml
- Changelog JSON feed: https://verifymcp.io/servers/com-zipirapp-zipir/api.json
- HTML version of this page: https://verifymcp.io/servers/com-zipirapp-zipir/api