# isGenAI

> Free read-only checks for evidence of AI use in text, links, images, saved prompts and C2PA.

- name: com.isgenai/checker
- version: 1.0.2
- connection class: R0 (autonomous)
- trust: 96/100
- quality: 88/100 (good)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/com.isgenai%2Fchecker
- tools: https://api.protogrid.dev/v1/servers/com.isgenai%2Fchecker/tools

**An agent can connect right now, no human step.** Remote endpoint, no authentication, reachable on the last probe.

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "checker": {
      "type": "http",
      "url": "https://isgenai.com/mcp"
    }
  }
}
```


## Trust

- hygiene: 100
- liveness: 100
- freshness: 100
- provenance: 85
- drivers: +repository +dns-namespace +website +reachable +uptime-100% +updated-23d-ago
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 88/100 (good)

### protocol: 75 (weight 25)

- warn `protocol.modern`: newest supported version is 2025-11-25; 2026-07-28 not supported, older versions are deprecated until 2027-07-28
- n/a `protocol.stateless`: only applies to 2026-07-28 servers
- pass `protocol.transport`: streamable HTTP
- n/a `protocol.list_ttl`: only applies to 2026-07-28 servers

### authorization: n/a (weight 15)

- n/a `auth.prm`, `auth.as_metadata`, `auth.cimd`: no remote uses OAuth
- n/a `auth.secret_in_url`: no templated URL

### tool hygiene: 100 (weight 30)

- pass `tools.descriptions`: every tool has a description
- pass `tools.description_length`: descriptions are concise
- pass `tools.schemas`: every tool has an object input schema
- pass `tools.annotations`: 6 of 6 tools declare readOnlyHint or destructiveHint
- pass `tools.directory_hints`: every tool declares readOnlyHint, destructiveHint, idempotentHint and openWorldHint
- pass `tools.token_cost`: about 1,685 tokens to load every tool
- pass `tools.api_dump`: tools are not a one-to-one API dump
- pass `tools.hidden_text`: no hidden text in tool names, descriptions, schemas or annotations

### stability: 83 (weight 15)

- warn `stability.changes`: 13 tool changes in 30 days
- pass `stability.rug_pull`: no tool changed its meaning under the same name
- pass `stability.new_links`: no tool change named a new site

### dependencies: n/a (weight 15)

- n/a `deps.known_vulns`, `deps.mcp_sdk_version`, `deps.resolvable`: no npm or PyPI package

- tools: 6, about 1,685 tokens to load them all
- tool set hash: 86fad6178a641c038f30c4e9e4977f9c
- tools last changed: 2026-11-09T00:18:57.258Z
- badge: [![protogrid quality](https://protogrid.dev/badge/com.isgenai/checker.svg)](https://protogrid.dev/servers/com.isgenai/checker)
- Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed.

### Tool changes

- 2026-11-09 `estimate_x_ai_replies` changed (title, annotations)
- 2026-11-09 `read_content_credentials` changed (title, annotations)
- 2026-11-09 `extract_saved_prompt` changed (title, annotations)
- 2026-11-09 `check_image` changed (title, description, annotations)
  - before: Check an image (PNG, JPEG, WebP, AVIF or single-page TIFF, up to 20 MiB) by public URL or base64 for C2PA Content Credentials, AI labels, saved generation settings and metadata. Evidence comes from records in the file, not a visual-only verdict. No public report is created. Docs: https://isgenai.com/developers/mcp
  - after: AI image detector: check an image (PNG, JPEG, WebP, AVIF or single-page TIFF, up to 20 MiB) by public URL or base64. Reads the evidence in the file first (C2PA Content Credentials naming the generator or camera, IPTC AI labels, prompts saved by Stable Diffusion tools, EXIF), then adds an AI-likelihood score from the image detector. Use for: is this photo real or AI, is this a deepfake, was this made with Midjourney, DALL-E, Firefly or ChatGPT? A score is an estimate; records are evidence. No public report is created. Docs: https://isgenai.com/developers/mcp
- 2026-11-09 `check_url` changed (title, description, annotations)
  - before: Fetch a public web page or file and read AI labels, AI-use statements, Content Credentials and saved creation records found there. Logins, paywalls and CAPTCHAs are not bypassed. No public report is created. Docs: https://isgenai.com/developers/mcp
  - after: Check a public web page or file URL for AI: fetches it and reads AI labels, AI-use statements, Content Credentials and saved creation records found there, with an AI-likelihood score for the image. Use for: is this article, news photo, product image or social post AI-generated or fake? Logins, paywalls and CAPTCHAs are not bypassed. No public report is created. Docs: https://isgenai.com/developers/mcp
- 2026-11-09 `check_text` changed (title, description, annotations)
  - before: Check pasted text or code (up to 16,000 characters) for statements of AI use and named AI tools such as ChatGPT, Claude, Gemini or Copilot, quoting the exact sentence. For 100+ characters it also returns an AI-likelihood estimate, labelled as an estimate, not proof of authorship. Private: no public report is created. Docs: https://isgenai.com/developers/mcp
  - after: AI text detector: check pasted text or code (up to 16,000 characters, 30 languages) for AI. Returns an AI-likelihood score with the reasons behind it (300+ characters, about 100 for Chinese, Japanese or Korean), flags chatbot leftovers such as ChatGPT citation tags and unfilled placeholders, and quotes any statement of AI use or named AI tool (ChatGPT, Claude, Gemini, Copilot) word for word. Use for: is this essay, email, post, review or code written by AI or ChatGPT? A score is an estimate, not proof of authorship. Private: no public report is created. Docs: https://isgenai.com/developers/mcp
- 2026-11-01 `estimate_x_ai_replies` added
  - Estimate what share of the replies under an X (Twitter) post look AI-generated or bot-driven. Pass the reply texts with any author details (handle, bio, followers, following, joined date, default avatar) and timestamps you have; the post text and time sharpen the estimate. Returns the percentage, a band and a 0-100 score with signals for every reply. An estimate from patterns, not proof about any account. Docs: https://isgenai.com/x-ai-replies
- 2026-11-01 `check_text` changed (description)
  - before: Check pasted text or code (up to 16,000 characters) for statements of AI use and named AI tools such as ChatGPT, Claude, Gemini or Copilot, quoting the exact sentence. Does not judge authorship from writing style. Private: no public report is created. Docs: https://isgenai.com/developers/mcp
  - after: Check pasted text or code (up to 16,000 characters) for statements of AI use and named AI tools such as ChatGPT, Claude, Gemini or Copilot, quoting the exact sentence. For 100+ characters it also returns an AI-likelihood estimate, labelled as an estimate, not proof of authorship. Private: no public report is created. Docs: https://isgenai.com/developers/mcp
- 2026-10-22 `read_content_credentials` changed (description, annotations)
  - before: Read C2PA Content Credentials from an image, video, audio or PDF file (by public URL or base64, up to 20 MiB). Reports the manifest, recorded actions and digital source type, with cryptographic validation and signer trust as separate findings.
  - after: Read C2PA Content Credentials from an image, video, audio or PDF file (by public URL or base64, up to 20 MiB). Reports the manifest, recorded actions and digital source type, with cryptographic validation and signer trust as separate findings. Docs: https://isgenai.com/developers/mcp
- 2026-10-22 `extract_saved_prompt` changed (description, annotations)
  - before: Read the prompt, negative prompt, model and settings saved inside an image by AUTOMATIC1111, ComfyUI (including Flux workflows), InvokeAI or NovelAI. Only reads records stored in the file; it never reconstructs or guesses a prompt. Screenshots and re-uploads usually remove these records.
  - after: Read the prompt, negative prompt, model and settings saved inside an image by AUTOMATIC1111, ComfyUI (including Flux workflows), InvokeAI or NovelAI. Only reads records stored in the file; it never reconstructs or guesses a prompt. Screenshots and re-uploads usually remove these records. Docs: https://isgenai.com/developers/mcp

Full history: https://api.protogrid.dev/v1/servers/com.isgenai%2Fchecker/changes

## Remotes

- https://isgenai.com/mcp (streamable-http, auth none, reachable true, uptime30d 1)

## Packages

_none_

## Tools (6)

- `check_image`: AI image detector: check an image (PNG, JPEG, WebP, AVIF or single-page TIFF, up to 20 MiB) by public URL or base64. Reads the evidence in the file first (C2PA Content Credentials naming the generator or camera, IPTC AI labels, prompts saved by Stable Diffusion tools, EXIF), then adds an AI-likelihood score from the image detector. Use for: is this photo real or AI, is this a deepfake, was this made with Midjourney, DALL-E, Firefly or ChatGPT? A score is an estimate; records are evidence. No public report is created. Docs: https://isgenai.com/developers/mcp
- `check_text`: AI text detector: check pasted text or code (up to 16,000 characters, 30 languages) for AI. Returns an AI-likelihood score with the reasons behind it (300+ characters, about 100 for Chinese, Japanese or Korean), flags chatbot leftovers such as ChatGPT citation tags and unfilled placeholders, and quotes any statement of AI use or named AI tool (ChatGPT, Claude, Gemini, Copilot) word for word. Use for: is this essay, email, post, review or code written by AI or ChatGPT? A score is an estimate, not proof of authorship. Private: no public report is created. Docs: https://isgenai.com/developers/mcp
- `check_url`: Check a public web page or file URL for AI: fetches it and reads AI labels, AI-use statements, Content Credentials and saved creation records found there, with an AI-likelihood score for the image. Use for: is this article, news photo, product image or social post AI-generated or fake? Logins, paywalls and CAPTCHAs are not bypassed. No public report is created. Docs: https://isgenai.com/developers/mcp
- `estimate_x_ai_replies`: Estimate what share of the replies under an X (Twitter) post look AI-generated or bot-driven. Pass the reply texts with any author details (handle, bio, followers, following, joined date, default avatar) and timestamps you have; the post text and time sharpen the estimate. Returns the percentage, a band and a 0-100 score with signals for every reply. An estimate from patterns, not proof about any account. Docs: https://isgenai.com/x-ai-replies
- `extract_saved_prompt`: Read the prompt, negative prompt, model and settings saved inside an image by AUTOMATIC1111, ComfyUI (including Flux workflows), InvokeAI or NovelAI. Only reads records stored in the file; it never reconstructs or guesses a prompt. Screenshots and re-uploads usually remove these records. Docs: https://isgenai.com/developers/mcp
- `read_content_credentials`: Read C2PA Content Credentials from an image, video, audio or PDF file (by public URL or base64, up to 20 MiB). Reports the manifest, recorded actions and digital source type, with cryptographic validation and signer trust as separate findings. Docs: https://isgenai.com/developers/mcp