wfgy-bbpf.md

tools	read write edit grep
arguments	$TOPIC

WFGY BBPF - Multi-Path Progression

Execute BBPF (BigBig Progressive Flow) for parallel reasoning exploration across multiple solution paths.

Based on the WFGY project: https://github.com/onestardao/WFGY

Formula

x_next = x + Σ(V_i(ε_i, C)) + Σ(W_j(dt, dO) * P_j)

Where:
- x: Current state vector
- V_i: Perturbation function for path i
- W_j: Dynamic weight for path j
- P_j: Importance/probability of path j
- ε_i: Creative perturbation factor
- C: Context constraints

Instructions

1. Initialize Multi-Path Exploration

   • Parse topic: "$TOPIC"
   • Load current state from .wfgy/context.json
   • Define N parallel paths (default N=5)
   • Set exploration parameters

2. Generate Reasoning Paths

   • For each path i (1 to N):
     • Apply creative perturbation V_i with variance ε_i
     • Maintain context constraints C
     • Generate unique reasoning trajectory
   • Calculate path probabilities P_j based on:
     • Semantic coherence
     • Logic consistency
     • Historical success rates

3. Weight Path Contributions

   • Compute dynamic weights W_j using:
     • Semantic tension: W_j = exp(-ΔS_j)
     • Temporal factor: dt (time delta)
     • Output difference: dO (divergence measure)
   • Normalize weights: Σ W_j = 1

4. Synthesize Next State

   • Combine all paths: x_next = x + Σ V_i + Σ(W_j * P_j)
   • Evaluate convergence criteria
   • Select top K paths for detailed exploration
   • Prune low-probability paths (P_j < 0.1)

5. Record Path Analysis

   • Save path explorations to .wfgy/paths/bbpf_session.json
   • Create semantic nodes for significant paths
   • Update .wfgy/context.json with new state
   • Log results in .wfgy/logs/bbpf.log

Output Format

BBPF Multi-Path Analysis
═══════════════════════════════
Topic: "$TOPIC"
Paths Explored: [N]

Path Rankings:
1. [Path Description] (P: [probability], W: [weight])
   Key Insight: [main finding]
   
2. [Path Description] (P: [probability], W: [weight])
   Key Insight: [main finding]
   
3. [Path Description] (P: [probability], W: [weight])
   Key Insight: [main finding]

Synthesis:
[Combined insights from weighted paths]

Recommended Path:
[Highest probability path with reasoning]

Alternative Paths:
[Other viable options with trade-offs]

Advanced Configuration

{
  "num_paths": 5,
  "perturbation_variance": 0.3,
  "pruning_threshold": 0.1,
  "convergence_criteria": 0.05,
  "max_iterations": 10
}

Use Cases

• Complex problem solving requiring multiple approaches
• Creative ideation and brainstorming
• Risk assessment through scenario exploration
• Decision making under uncertainty