Figma Context MCP
Give your coding agent access to your Figma data. Implement designs in any framework in one-shot.
作者による説明(英語の原文)
Detect grooming, bullying, fraud, and 16+ online threats across text, voice, image, and video.
この3つを1つのスコアにまとめることはしません。意見、測定、人気はそれぞれ別のものだからです。
このサーバーのやさしい解説はまだ書いていません。以下はレジストリから読み取った事実です。
使っているAIアプリを選んで、設定をコピーしてください。
TUTELIQ_API_KEY「秘密」と表示された値は、あなたのアカウントの鍵です。絶対に人に教えたり、チャットに貼り付けたりしないでください。
必要なもの
どのアプリを使っていますか?
上でアプリを選んでください。選んだアプリを覚えておくので、次からはボタンに「…にインストール」と表示されます。
ウェブ、デスクトップ、モバイルのClaudeで使えます:
https://api.tuteliq.ai/mcpClaude Desktop → Settings → Developer → Edit Config を開き、claude_desktop_config.json にこれを貼り付けます
{
"mcpServers": {
"tuteliq": {
"command": "npx",
"args": [
"-y",
"@tuteliq/mcp"
],
"env": {
"TUTELIQ_API_KEY": "your-value"
}
}
}
}ターミナルでこれを実行します
claude mcp add --transport http tuteliq https://api.tuteliq.ai/mcpclaude mcp add tuteliq --env TUTELIQ_API_KEY=your-value -- npx -y @tuteliq/mcp動かす前に、空欄の値を入力してください。
プロジェクトの .cursor/mcp.json、または ~/.cursor/mcp.json に貼り付けます
{
"mcpServers": {
"tuteliq": {
"url": "https://api.tuteliq.ai/mcp"
}
}
}プロジェクトの .cursor/mcp.json、または ~/.cursor/mcp.json に貼り付けます
{
"mcpServers": {
"tuteliq": {
"command": "npx",
"args": [
"-y",
"@tuteliq/mcp"
],
"env": {
"TUTELIQ_API_KEY": "your-value"
}
}
}
}ブラウザがアプリを開いてよいか確認し、そのあとアプリが設定を表示して確認を求めます。キーの値は空欄のままなので、自分で入力してください。
プロジェクトの .vscode/mcp.json に貼り付けます
{
"servers": {
"tuteliq": {
"type": "http",
"url": "https://api.tuteliq.ai/mcp"
}
}
}プロジェクトの .vscode/mcp.json に貼り付けます
{
"servers": {
"tuteliq": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@tuteliq/mcp"
],
"env": {
"TUTELIQ_API_KEY": "your-value"
}
}
}
}ブラウザがアプリを開いてよいか確認し、そのあとアプリが設定を表示して確認を求めます。キーの値は空欄のままなので、自分で入力してください。
ChatGPTがつなげるのは、インターネット上のサーバーだけです:
https://api.tuteliq.ai/mcpメニューの名前は、プランや時期によって少し違うことがあります。
選んだアプリは、この端末のCookieに保存されます。アカウントは不要です。
この設定は、作者のレジストリ登録情報から作成しています。READMEと違う場合は、READMEに従ってください。
このサーバーは、ツール一覧を表示する前にサインインを求めます。
誰でも見られる情報をチェックしており、コードは読んでいません。当てはまらない項目は数に入れていません。 · 2026年11月7日にチェック
レジストリ名はドメイン tuteliq.ai で確認済みです(レジストリがドメインの所有を確認しています)
このリポジトリはまだGitHubから読み取っていません
このリポジトリはまだGitHubから読み取っていません
274ミリ秒で応答(HTTP 401)
OAuthでのサインインに対応しているため、AIアプリがログイン画面へ案内してくれます
まだチェックしていません。エンドポイントをチェックするときにツール一覧を読み取ります
レジストリでこのカテゴリにある有効なすべてのサーバーと比較しています。スターは、公開者がリポジトリの所有者である場合だけ数えます。
当てはまる基本チェック3項目のうち3項目に合格しました。コードは読んでいません。キーやトークンを追加する前に、サーバーがどんなアクセスを求めているかを確認し、必要な権限だけを持つキーを使ってください。
入力が必要なもの:TUTELIQ_API_KEY
どちらでも使えます。パソコンにインストールしても、クラウド版につないでもかまいません。
GitHubリポジトリの最終更新は–、最新バージョン(4.2.6)がレジストリに届いたのは5 日前です。
Claude Desktop、Claude Code、Cursor、VS CodeなどのMCP対応アプリで使えます。URLでつながるので、リモートMCPサーバーに対応したアプリ(開発者モードのChatGPTなど)でも使えます。
同じカテゴリには:Figma Context MCP, Claude Real Video, Figma MCP Server
同じカテゴリで、ほかの作者が作ったサーバー。スターの多い順。
カードの見方:★ GitHubスターi3つの柱:レビュー · チェック · 利用状況i更新=最新バージョンをレジストリに公開した日i
Give your coding agent access to your Figma data. Implement designs in any framework in one-shot.
Let any LLM watch a video locally — and search everything it has ever watched.
The Figma MCP server brings Figma design context directly into your AI workflow.
作者のリポジトリから書かれたとおりにコピーしたものです。PopNx MCPは内容を確認・推奨していません。HTML、画像、スクリプトは取り除いています。リンクはこのサイトの外へ移動します。
Tuteliq MCP Server
MCP server for Tuteliq - AI-powered child safety tools for Claude
API Docs • Dashboard • Trust • Changelog • Discord
Tuteliq MCP Server brings AI-powered child safety tools directly into Claude, Cursor, and other MCP-compatible AI assistants. Ask Claude to check messages for bullying, detect grooming patterns, or generate safety action plans.
Reads context, not just keywords. Every detector understands coded slang, emoji, leetspeak, algospeak, and deliberate filter evasion, and weighs the conversation around a message — so it tells gaming trash-talk apart from targeted harassment instead of drowning your team in false positives. This coded-language resilience is platform-wide (it applies to grooming, fraud, radicalisation, and the rest, not just bullying) and is built from our own research into how bad actors evade moderation. In an internal benchmark of coded-language and filter-evasion cases, Tuteliq detected roughly 1.7x more of them than leading general-purpose moderation APIs (319-case evasion set; vendors unnamed).
Fast mode. Pass verdict_only: true on detect_grooming or detect_bullying to get just the verdict (risk level, flags, recommended action) without the per-message breakdown — lower latency for real-time screening. The verdict itself is unchanged.
Interactive results. In hosts that support MCP Apps, results render as interactive cards rather than walls of JSON — see Interactive widgets below.
Twelve widgets return a rendered card instead of raw text in hosts that support MCP Apps (Claude desktop and web, and other MCP-compatible clients). Everywhere else the same data arrives as structuredContent, so nothing depends on the UI.
Every card carries the same frame: a chrome bar naming the tool that produced the result, the result itself, and a footer with the data-handling note and a Trust Center link. In a transcript holding a dozen results, the chrome bar is what tells you which is which.
| Widget | Tools |
|---|---|
| Detection result | detect_bullying, detect_grooming, detect_unsafe, analyze, and the other detect_* tools |
| Multi-endpoint | analyse_multi |
| Emotions | analyze_emotions |
| Media | analyze_voice, analyze_image, analyze_video, analyze_document |
| Synthetic media | detect_synthetic_text, detect_synthetic_image, detect_synthetic_audio, detect_synthetic_video |
| Action plan | get_action_plan |
| Incident report | generate_report |
| Incidents overview | get_incidents_overview |
| Incidents list | list_incidents |
| Incident detail | get_incident |
| Incident trends | get_incident_trends |
| Moderation queue | moderation_queue |
Severity is rankable by colour. The ramp runs monotonically from safe to critical, so two chips can be compared without reading their labels:
| Level | Colour | |
|---|---|---|
critical | #9C3A29 | |
high | #C2543A | |
medium | #D98A3D | |
low | #B7C2D4 | |
safe | #19B79A |
Design notes. The widgets are deliberately calm. They report on grooming, self-harm, and abuse, and a card that animates or pulses at the reader turns material that is already distressing into an alarm they cannot dismiss. Severity is carried by a rule and a glyph, not by motion. The crisis-support card leads with reassurance rather than the severity colour, and its helpline numbers are the largest targets on the card because transcribing digits under stress is where people fail.
Widgets are read-only renderers by design. Selecting incidents in the list widget assembles the ID list for a batch_review_incidents call you fire yourself — the mutating call still goes through your host's approval step, so the human-in-the-loop stays in the loop.
npm run preview:ui # builds every widget against fixture data
open dist-preview/__preview.html
Widget source lives in ui/src. Design tokens are centralised in ui/src/theme.ts; prefer them over colour literals so the palette stays in one place.
| Tool | Description |
|---|---|
detect_bullying | Analyze text for bullying, harassment, and gaming toxicity — including coded slang, emoji, and deliberate filter evasion, with context that tells trash-talk apart from genuine harm |
detect_grooming | Detect grooming patterns and predatory behavior in conversations |
detect_unsafe | Identify unsafe content (self-harm, violence, explicit material) |
analyze | Quick comprehensive safety check (bullying + unsafe) |
analyse_multi | Run multiple detection endpoints on a single piece of text in one call |
batch_analyze | Analyze up to 50 items in a single request — all twelve detection types (bullying, grooming, unsafe, emotions, social_engineering, app_fraud, romance_scam, mule_recruitment, gambling_harm, coercive_control, vulnerability_exploitation, radicalisation) — ideal for bulk triage |
analyze_emotions | Analyze emotional content and mental state indicators — accepts single text or full conversations |
get_action_plan | Generate age-appropriate guidance for safety situations |
generate_report | Create incident reports from conversations |
| Tool | Description |
|---|---|
detect_social_engineering | Detect social engineering tactics (pretexting, urgency fabrication, authority impersonation) |
detect_app_fraud | Detect app-based fraud (fake investment platforms, phishing apps, subscription traps) |
detect_romance_scam | Detect romance scam patterns (love-bombing, financial requests, identity deception) |
detect_mule_recruitment | Detect money mule recruitment tactics (easy-money offers, bank account sharing) |
detect_gambling_harm | Detect gambling-related harm indicators (chasing losses, concealment, distress) |
detect_coercive_control | Detect coercive control patterns (isolation, financial control, monitoring, threats) |
detect_vulnerability_exploitation | Detect exploitation of vulnerable individuals (elderly, disabled, financially distressed) |
detect_radicalisation | Detect radicalisation indicators (extremist rhetoric, us-vs-them framing, ideological grooming) |
| Tool | Description |
|---|---|
analyze_voice | Transcribe audio and run safety analysis on the transcript |
analyze_image | Analyze images for visual safety + OCR text extraction |
analyze_video | Analyze video files for safety concerns via key frame extraction (supports mp4, mov, avi, webm, mkv) |
analyze_document | Analyze PDF documents for safety concerns — per-page multi-endpoint detection with chain-of-custody hashing (max 50MB, 100 pages) |
| Tool | Description |
|---|---|
detect_synthetic_text | Detect AI-generated text across 10 child-safety categories (synthetic CSAM, deepfake scripts, AI grooming) |
detect_synthetic_image | 6-signal forensic pipeline: vision AI, EXIF metadata, pixel stats, C2PA Content Credentials, watermarks, pHash |
detect_synthetic_audio | Dual-signal forensics: transcript + mel spectrogram vision + quantitative audio statistics |
detect_synthetic_video | 5-track analysis: per-frame vision, temporal face consistency, lip-sync correlation, spectral audio, transcript |
| Tool | Description |
|---|---|
create_verification_session | Create a session for age or identity verification — returns a URL for the user to complete the flow |
get_verification_session | Poll session status — returns full document intelligence (MRZ, barcode, authenticity, face match, liveness) |
cancel_verification_session | Cancel an active session (no credits consumed) |
These three tools are registered when you run this package yourself. The hosted connector does not serve them.
Read the incident store, triage a queue, and record moderator decisions. The review tools emit signed receipts for EU AI Act Art 14 human-oversight evidence.
| Tool | Description |
|---|---|
get_incidents_overview | Counts by category, severity, source, status and platform over a window |
list_incidents | Paginated, filterable incident list |
get_incident | Full detail for one incident, including the risk trajectory across messages |
get_incident_trends | Incident volume bucketed by hour, day or week, split by severity |
moderation_queue | Moderator triage console: the unreviewed queue, the next item, and — optionally — your own analysis trace and recommended decision, rendered for human sign-off. Read-only |
review_incident | Record a moderator decision (confirm / downgrade / escalate / reclassify / dismiss) with a signed receipt |
batch_review_incidents | Apply one decision across many incidents in a single call |
get_audit_receipt | Fetch the signed receipt for a past inference |
get_audit_logs | Query the audit log |
The decision is the moderator's, and the card makes them take it. The action buttons call review_incident through the host, because a moderator clicking "Escalate" *is* the human decision. They do not fire on one click: review_incident persists an override and emits a signed Art 12 audit receipt and requires a reason_code, so the button opens a reason picker and a second click commits. Nothing is ever defaulted into that receipt on the moderator's behalf.
The reasoning, confidence and analysis trace on the card are supplied by the calling assistant and are labelled as such — an argument for a human to weigh, not a Tuteliq measurement.
Pass operator_name to brand the header with the customer or team name. Omit it and the card is unbranded — it is never defaulted to a placeholder.
| Tool | Description |
|---|---|
list_webhooks | List all configured webhooks |
create_webhook | Create a new webhook endpoint |
update_webhook | Update webhook configuration |
delete_webhook | Delete a webhook |
test_webhook | Send a test payload to verify webhook |
regenerate_webhook_secret | Regenerate webhook signing secret |
| Tool | Description |
|---|---|
get_pricing | Get available pricing plans |
get_pricing_details | Get detailed pricing with features and limits |
| Tool | Description |
|---|---|
get_usage_history | Get daily usage history |
get_usage_by_tool | Get usage by tool/endpoint |
get_usage_monthly | Get monthly usage with billing info |
get_usage_summary | Get today's usage: requests made, succeeded and failed, remaining today, and the rate limit |
get_usage_quota | Get real-time rate-limit status — pre-flight check before batch runs |
冒頭だけを表示しています。続きはリポジトリで読めます