X Algorithm MCP

Score, optimize, and generate tweets using theactual signal weightsfrom X's open-source recommendation algorithm.

This toolkit reverse-engineers X's recommendation algorithm from theopen-source codebaseand gives you:

# Clone git clone https://github.com/MrChartist/x-algorithm-toolkit.git cd x-algorithm-toolkit # Install & Run npm install npm run dev

{ "mcpServers": { "x-algorithm": { "command": "npx", "args": ["-y", "x-algorithm-mcp"] } } }

cd mcp && npm install && npm run build

11 Tools:score_tweet,check_filters,analyze_hook,get_signals,get_optimal_schedule,compare_tweets,detect_spam_patterns,diversity_penalty,classify_niche,get_thread_strategy,full_audit

3 Resources:x-algorithm://weights,x-algorithm://filters,x-algorithm://niches

400+ lines of domain expertise that makes any AI an X algorithm expert:

mkdir -p ~/.gemini/antigravity/skills/x-algorithm-expert curl -o ~/.gemini/antigravity/skills/x-algorithm-expert/SKILL.md \ https://raw.githubusercontent.com/MrChartist/x-algorithm-toolkit/master/skill/SKILL.md

- All 22 signal weights with optimization strategies
- Phoenix Grok-based transformer architecture
- 77+ topic categories fromtopic_ids_filter.rs
- 10-point scoring system with letter grades
- Hook writing rules and 7 templates
- Author diversity decay formula
- OON viral mechanics
- Thread building strategy (6 templates)
- 18 safety filter checklist
- Niche-specific optimization (7 niches)

x-algo-toolkit/ ├── src/ # React Web App (13 tabs) │ ├── engine/ # Scoring engine + AI abstraction │ └── components/ # 15 UI components ├── mcp/ # MCP Server (11 tools, 3 resources) │ ├── src/ # TypeScript source │ ├── dist/ # Compiled output │ └── package.json # npm: x-algorithm-mcp ├── skill/ # AI Skill │ └── SKILL.md # 400+ line expertise file └── public/ # Static assets

- Frontend: React 19, TypeScript, Vite
- Animations: Framer Motion
- AI: Multi-provider (Groq/Gemini/OpenAI/Claude)
- MCP:@modelcontextprotocol/sdkv1.12+
- Storage: localStorage (no backend needed)

Data sourced from X's open-source algorithm:weighted_scorer.rs,ranking_scorer.rs,author_diversity_scorer.rs,oon_scorer.rs,topic_ids_filter.rs,recsys_model.py

Free MCP that drives an audit of your marketing. Your AI connects, adsOS digs through your ads, email and site, and hands back a growth plan you can run today.

Marketing insights and audience analysis from Audiense reports, covering demographic, cultural, influencer, and content engagement analysis.

AI-powered social media profile analysis, engagement scoring, trend detection, and hashtag research.

Connect AI agents to project data, analytics, research, content planning, media generation, and growth workflows.

Perplexity for social: ask what competitors are running across social and get answers grounded in real ad and creative data, not guesses.

Official MCP server for Amazon product research, competitor analysis, keyword discovery, and PPC planning using real Amazon marketplace data.

Publish, schedule, moderate your social inbox, and pull analytics across every major network directly from any AI assistant, via Eclincher.

With Epovest, businesses make AIs recommend them: your market's real questions put to the engines, every answer archived with its cited sources and dated, and the levers to act on them, over a hosted Streamable HTTP endpoint.

Live TikTok creator intelligence MCP — 7 tools (analyze creators, score hooks, remix scripts, predict virality) that chain automatically via a recommended_chain that pre-fills the next 3 tool calls.

Instagram Graph API server for Business/Creator accounts — 24 tools for posting, comments, DMs, and insights.

Connect Claude or ChatGPT to your LinkedIn account to see who engaged on your posts, track performance and profile viewers, spot warm outreach signals, and publish content — built for B2B marketers and creators.