feature/genetic-symbolic-reflexive

[DarshanKumar89]

Pull Request

Description

This PR implements the advanced modular memory and agentic architecture as outlined in Issue #32.
It introduces a full symbolic memory stack (GraphMemoryAgent, MemoryTripleStore, Deduplicator, Scorer, Timeline), advanced reflexive agents (JudgeAgent, PlannerAgent, RewriterAgent, ThinkerAgent, SelfReflectAgent, FactExtractorAgent), evolutionary modules (MetaControllerAgent, AgentMutator, AgentArena, MultiObjectiveJudge, EvolutionMemory), and a modular pipeline runner.
Additional advanced features include semantic deduplication, contradiction detection, memory merging, timeline queries, and hybrid evolutionary-reflexive agent demos.

Fixes: #32

Type of Change

• [x] New feature
• [x] Documentation update
• [x] Refactoring
• [ ] Bug fix
• [ ] Other (please describe):

Checklist

• [x] My code follows the style guidelines of this project
• [x] I have performed a self-review of my code
• [x] I have commented my code, particularly in hard-to-understand areas
• [x] I have made corresponding changes to the documentation
• [x] I have added tests that prove my feature works
• [x] New and existing unit tests pass locally with my changes
• [x] I have checked my code and corrected any misspellings

Screenshots (if applicable)

N/A

Additional Context

• Implements and exceeds the modular roadmap in Issue #32.
• Adds advanced demos and developer onboarding examples.
• All modules are pluggable, extensible, and developer-friendly.
• See examples/evolutionary/advanced_graph_memory_integration_demo.py for a showcase of new features.

[DarshanKumar89]

Related : https://github.com/multimindlab/multimind-sdk/issues/36

🟢 Developed Features Summary

1. True DAG-Based Orchestration

• Agents and their dependencies are modeled as nodes and edges in a true Directed Acyclic Graph (DAG) using networkx.
• Supports both linear and complex, branching workflows.

2. YAML-to-DAG Pipeline Definition

• Pipelines can be defined declaratively in YAML.
• YAML parser (build_dag_from_yaml) converts YAML to an executable DAG.
• Supports:
  • Linear flows
  • Parallel branches
  • Conditional routers/fallbacks
  • Performance-aware and symbolic evolution patterns

3. Parallel and Conditional Execution

• DAG engine executes independent nodes in parallel.
• Supports conditional branches and router nodes (e.g., if, switch, fallback).

4. Runtime Graph Mutation

• DAG can be mutated at runtime (nodes/edges can be added, removed, or replaced).
• MetaControllerAgent enables dynamic adaptation based on runtime feedback or metrics.

5. Performance-Aware & Symbolic Evolution

• Example mutation policies for:
  • Inserting nodes if a metric (e.g., latency) is high
  • Removing nodes if a metric is low
  • Replacing nodes dynamically

6. Integration with Real-World Agents

• Runners and examples show how to register and use real agent classes (e.g., ThinkerAgent, SelfReflectAgent) in YAML-defined pipelines.

7. Visualization Tools

• Utility to visualize any DAG pipeline as a graph using matplotlib.
• Script to visualize YAML-defined pipelines.

8. Example Suite

• Comprehensive set of YAML pipeline examples:
  • Linear, branching, conditional, performance-adaptive, and symbolic evolution workflows
• Runner scripts for each example, demonstrating both mock and real agent integration

9. Documentation

• README and quickstart guide for YAML-to-DAG orchestration.
• Example code and usage instructions for all major features.

---