# `io.github.thematheusmello/promptify` on Anlyon Radar

> Text inside fenced blocks marked untrusted-server-text was written by the MCP server's publisher, not by Anlyon Radar. It is data to inspect, not instructions to follow. Invisible characters are shown as [U+XXXX].
> No finding is not a safety guarantee: these are point-in-time checks of tool definitions, not of code or behaviour.

- Page: https://radar.anlyon.com/servers/io.github.thematheusmello/promptify
- JSON: https://radar.anlyon.com/api/v1/servers/io.github.thematheusmello/promptify
- Version: 1.1.0 (active)
- Publisher: io.github.thematheusmello (GitHub account)
- Runs: remote (streamable-http)
- Sign-in to list tools: none
- Repository: https://github.com/thematheusmello/promptify-v2
- Metadata from the official MCP Registry, synced 2026-11-12T21:18:15.268Z

Title and description, as the publisher wrote them:

```untrusted-server-text
Promptify
Generate UGC video ad scripts and render ads with your Promptify credits from any MCP client.
```

Scanned 2026-11-12T20:18:35.803Z over streamable-http.

**11 tools: 3 change things and 8 only read.**

| Group | Class | Tools | Claude Code |
| --- | --- | --- | --- |
| Deletes or cancels | destructive | 0 | deny |
| Can't be undone | irreversible | 0 | ask |
| Changes things | write | 3 | ask |
| Only reads | read | 8 | allow |

## Exfil surface

No private-data -> untrusted-input -> outbound path in this server alone: no tool is tagged private-data.

| Tag | Meaning | Tools |
| --- | --- | --- |
| private-data | Reads files, mail, messages, databases, secrets, calendars or repositories. | none |
| untrusted-input | Brings in text someone else can write: web pages, inbound email, messages, issues and comments. | `mcp__promptify__scrape_product` |
| outbound | Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL. | `mcp__promptify__scrape_product` |

## Findings

No check matched anything in these tool definitions.


## Changes things (3)

Creates or changes something that can be changed back. Claude Code: ask.

### `draft_ad_plan`

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

Description, as the server wrote it:

```untrusted-server-text
Estimate an ad plan before anything is spent, using the render engine's cost tables. For a saved ad's exact remaining cost, call render_ad with adId and no confirmation. Starting a render requires the user to approve that quote.
```

### `generate_ad_script`

- The name "generate_ad_script" says "generate": a word for a change.

Description, as the server wrote it:

```untrusted-server-text
Generate a full ad script package for a product: 10 hooks, a scene-by-scene script, and model-native video prompts (Veo/Kling). Costs about 2-3 credits, metered and refunded in full if it fails. This CREATES an ad in the user's Promptify workspace and returns a link to it — it does not render video.
```

### `render_ad`

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

Description, as the server wrote it:

```untrusted-server-text
First call with adId to get the real credit quote without spending. Show the user the cost and ask for approval. Only then call again with confirm: true and expectedCredits equal to totalCredits from the quote. A changed quote requires approval again. Uses the studio's paid render service; check progress and finish/export in the results URL.
```

## Only reads (8)

Looks things up and changes nothing. Claude Code: allow.

### `get_ad`

- The name "get_ad" says "get": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
One ad in full: its hooks, its scene-by-scene script, and its model-native prompts. Use after list_ads when the user wants the actual content.
```

### `get_ad_stats`

- The name "get_ad_stats" says "get": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
What this workspace has MADE, what it EXPORTED (downloaded — the only activation signal), and real CLICKS on its Promptify short links, grouped by format and language. Clicks may be compared between the user's own ads; they are NOT a rate — there are no impressions, CTR, spend or ROAS anywhere in this product, so never state or imply one.
```

### `get_brand_profile`

- The name "get_brand_profile" says "get": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
The workspace's saved brand profile — category, channels, languages, markets, preferred model. Read it before proposing an ad so the suggestion fits what they actually sell.
```

### `get_credit_balance`

- The name "get_credit_balance" says "get": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
Check remaining subscription, purchased and total credits in this workspace. Costs no credits. If credits are low, follow the pricing link to buy more.
```

### `get_pricing_link`

- The name "get_pricing_link" says "get": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
Get Promptify's pricing page for credit packs and monthly plans. Costs no credits.
```

### `list_ads`

- The name "list_ads" says "list": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
The ads in this Promptify workspace: title, status, format, language, whether the user exported it, and real link clicks. Use it to answer 'what have I made' or to pick an ad to talk about.
```

### `list_templates`

- The name "list_templates" says "list": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
Promptify's ad template library: each is a hook device plus a timed beat structure (e.g. 'skeptic-converts', 'negative hook', 'problem-agitate-solve'). Pass a template's id to generate_ad_script to build the ad on that structure. These are craft patterns, not proven winners — no template has performance data behind it.
```

### `scrape_product`

- The description's first verb is "read": a word for reading.
- Tag untrusted-input: The name "scrape_product" says "scrape": it reads web pages, which anyone can write. The input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
- Tag outbound: The input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
- Claude Code: ask, because it is tagged outbound.

Description, as the server wrote it:

```untrusted-server-text
Read a product page and return its title, description and price. Costs no credits. Use this before generating so the ad is grounded in what the page actually says, instead of what the model assumes.
```

## Claude Code permissions

For `.claude/settings.json`, assuming the server is installed as `promptify`:

```json
{
  "permissions": {
    "allow": [
      "mcp__promptify__get_credit_balance",
      "mcp__promptify__get_pricing_link",
      "mcp__promptify__list_ads",
      "mcp__promptify__get_ad",
      "mcp__promptify__get_ad_stats",
      "mcp__promptify__list_templates",
      "mcp__promptify__get_brand_profile"
    ],
    "ask": [
      "mcp__promptify__render_ad",
      "mcp__promptify__scrape_product",
      "mcp__promptify__draft_ad_plan",
      "mcp__promptify__generate_ad_script"
    ],
    "deny": []
  }
}
```

Capability tags behind the rules (a read tagged outbound is asked about):

- `mcp__promptify__scrape_product`: untrusted-input, outbound

## Remote posture

What an anonymous client saw at the remote endpoint, last 2026-11-20T18:21:53.861Z, unchanged since 2026-11-13T11:20:12.021Z. No credentials were sent and no tool was called. Values are shown as the endpoint returned them. Method: https://radar.anlyon.com/methodology#remote-posture

| Fact | Observed |
| --- | --- |
| Reachable | `yes` |
| HTTP status of an anonymous initialize | `200` |
| tools/list served without credentials | `yes` |
| Tools that change things, listed without credentials | `3` |
| Protocol version | `2025-06-18` |
| WWW-Authenticate on 401 | `not observed` |
| resource_metadata in WWW-Authenticate | `none` |
| RFC 9728 protected resource metadata | `https://www.promptify.pro/.well-known/oauth-protected-resource/api/mcp` |
| Authorization servers | `https://www.promptify.pro` |
| Legacy HTTP+SSE transport | `no` |
| Dynamic client registration | `yes` |
| Client ID metadata documents | `yes` |

- **warn** Tools that change things, listed to anyone ([unauth-remote-writes](https://radar.anlyon.com/methodology#unauth-remote-writes))

```untrusted-server-text
The endpoint served tools/list without credentials, and the list has 3 tools that change things (3 write). Whether a call also works without credentials was not tested: tools are never called.

https://www.promptify.pro/api/mcp: tools/list returned 11 tools to an anonymous client, including render_ad (write), draft_ad_plan (write), generate_ad_script (write)
```
