canvas-lms-mcp: Skill for Cursor

Bayesian Reasoning Agents

Three-agent system for hypothesis-driven development using Bayesian reasoning and git worktrees.

Architecture Overview

┌─────────────────────────────────────────────────────────────────────────┐
│                        Bayesian Orchestrator                             │
├───────────────────┬─────────────────────┬───────────────────────────────┤
│   Agent 1:        │   Agent 2:          │   Agent 3:                    │
│   Hypothesis      │   Evidence          │   Backwards                   │
│   Generator       │   Evaluator         │   Reasoner                    │
├───────────────────┼─────────────────────┼───────────────────────────────┤
│ - Hypotheses      │ - Evaluate evidence │ - P(Cause|Effect)             │
│ - Experiments     │ - Update beliefs    │ - Conditional variables       │
│ - Predictions     │ - Verify results    │ - Causal graphs               │
│ - WT Schemas      │ - Create worktrees  │ - Suggest hypotheses          │
└───────────────────┴─────────────────────┴───────────────────────────────┘
         │                    │                        │
         ▼                    ▼                        ▼
   worktree_schemas/    worktrees/branch-*     Causal Network

Quick Start

uv add numpy scipy pydantic

from agents.bayesian import (
    BayesianOrchestrator,
    OrchestratorConfig,
)

# Initialize orchestrator with all three agents
orchestrator = BayesianOrchestrator()

# Set up causal network for backwards reasoning
orchestrator.setup_causal_network(
    variables=[
        {"name": "bug_in_code", "description": "A bug exists in the code"},
        {"name": "test_failure", "description": "Tests are failing"},
        {"name": "recent_change", "description": "Recent code change"},
    ],
    causal_links=[
        {"cause": "bug_in_code", "effect": "test_failure", "strength": 0.8},
        {"cause": "recent_change", "effect": "bug_in_code", "strength": 0.6},
    ]
)

# Reason from observation
result = orchestrator.reason_from_observation(
    observation="Tests started failing after deployment",
    observation_vars={"test_failure": "true"}
)

print(result["hypotheses"])