graph LR
    A[Input samples dx/dy/dt] --> B[InputTrajectory]
    B --> C[MotorScorer.extract_features]
    C --> D[MotorFeatures<br/>noise_ratio, correction_rate, smoothness]
    D --> E[MotorScorer.score]
    E --> F[MotorScore<br/>human_score, ai_score, verdict]
    F --> G[CLI / API / MCP]

pip install humanproof
pip install "humanproof[api]"   # + FastAPI server
pip install "humanproof[mcp]"   # + MCP server for Claude

from humanproof import InputSample, InputTrajectory, MotorScorer

samples = [InputSample(dx=3.0, dy=2.0, dt=10.0) for _ in range(20)]
traj = InputTrajectory(samples=samples)
scorer = MotorScorer()
result = scorer.score(traj)
print(result.verdict, result.human_score)

humanproof score trajectory.json
humanproof batch ./trajectories/
humanproof log
humanproof status

pip install "humanproof[api]"
uvicorn humanproof.api:app --reload

curl -X POST http://localhost:8000/score -H 'Content-Type: application/json' \
  -d '{"samples": [{"dx":1,"dy":1,"dt":10}]}'

{
  "mcpServers": {
    "humanproof": {
      "command": "humanproof-mcp"
    }
  }
}

import json, openai
tools = json.load(open("tools/openai-tools.json"))
response = openai.chat.completions.create(
    model="gpt-4o",
    tools=tools,
    messages=[{"role": "user", "content": "Is this input human?"}]
)

- uses: sandeep-alluru/humanproof@v0.1.0
  with:
    trajectory-file: replay.json

humanproof/
├── src/humanproof/       # library source
│   ├── trajectory.py     # InputSample, InputTrajectory
│   ├── scorer.py         # MotorFeatures, MotorScore, MotorScorer
│   ├── store.py          # SQLite persistence
│   ├── report.py         # Rich / JSON / Markdown output
│   ├── cli.py            # Click CLI
│   ├── api.py            # FastAPI server
│   └── mcp_server.py     # MCP server
├── tests/                # 77 pytest tests, 95% coverage
├── examples/
│   ├── demo.py                          # end-to-end demo
│   ├── game_anticheat.py                # game anti-cheat integration example
│   ├── esports_integrity_monitor.py     # esports session integrity monitor
│   └── claude_computer_use_detection.py # Claude computer-use AI detection
├── docs/                 # 11-page MkDocs site
└── tools/openai-tools.json

from humanproof import InputSample, InputTrajectory, MotorScorer

# Human player trajectory — realistic noise, varied timing (dt 8–12ms)
human_deltas = [
    (3.1, 2.4, 9.0), (-1.2, 3.8, 11.0), (4.7, -0.9, 8.0), (2.3, 5.1, 10.0),
    (-0.8, 2.7, 12.0), (5.2, -1.4, 9.0), (1.9, 4.3, 10.0), (-2.6, 0.8, 8.0),
    (3.8, -3.1, 11.0), (0.4, 6.2, 9.0), (-1.7, 2.9, 10.0), (4.1, 0.3, 12.0),
    (2.8, -2.2, 8.0), (-0.5, 4.8, 10.0), (3.4, 1.7, 9.0), (1.1, -3.6, 11.0),
    (5.0, 2.1, 10.0), (-2.9, 3.5, 8.0), (0.7, -1.8, 12.0), (4.4, 2.6, 9.0),
]
human_samples = [InputSample(dx=dx, dy=dy, dt=dt) for dx, dy, dt in human_deltas]

# AI bot trajectory — unnaturally smooth, perfectly consistent timing (dt=16ms exactly)
bot_deltas = [
    (2.0, 2.0, 16.0), (2.0, 2.0, 16.0), (2.0, 2.0, 16.0), (2.0, 2.0, 16.0),
    (2.0, 2.0, 16.0), (2.0, 2.0, 16.0), (2.0, 2.0, 16.0), (2.0, 2.0, 16.0),
    (2.0, 2.0, 16.0), (2.0, 2.0, 16.0), (2.0, 2.0, 16.0), (2.0, 2.0, 16.0),
    (2.0, 2.0, 16.0), (2.0, 2.0, 16.0), (2.0, 2.0, 16.0), (2.0, 2.0, 16.0),
    (2.0, 2.0, 16.0), (2.0, 2.0, 16.0), (2.0, 2.0, 16.0), (2.0, 2.0, 16.0),
]
bot_samples = [InputSample(dx=dx, dy=dy, dt=dt) for dx, dy, dt in bot_deltas]

scorer = MotorScorer()

human_result = scorer.score(InputTrajectory(samples=human_samples))
bot_result   = scorer.score(InputTrajectory(samples=bot_samples))

print(f"[Player]  verdict={human_result.verdict}  human_score={human_result.human_score:.2f}  ai_score={human_result.ai_score:.2f}")
print(f"[Bot]     verdict={bot_result.verdict}  human_score={bot_result.human_score:.2f}  ai_score={bot_result.ai_score:.2f}")

if bot_result.verdict == "AI":
    print("\nFLAGGED: Suspected aimbot detected — trajectory referred to tournament integrity committee.")