Skip to main content

Agent SDK Overview

Benchmarked against: Anthropic — Agent SDK Overview Architecture: Dual Agentic Lines — Claude (30%) + LangGraph (70%) Principle: Both lines are truly agentic — autonomous perceive-reason-act-learn systems

SuperPortia operates two parallel agentic lines. This is not "pick one" — both run simultaneously with mutual disaster recovery.


Dual Agentic Lines


Line comparison

AspectLine 1: Claude AgenticLine 2: LangGraph
Share~30% of operations~70% of operations
StackClaude Code + Agent SDKLangGraph + LangChain + MCP Adapters
RoleArchitecture, decisions, code opsResearch, analysis, classification, patrol
EngineClaude Opus / Sonnet / HaikuGroq / Gemini / DeepSeek / Mistral / Zhipu
Tool accessAll MCP servers + built-in toolsMCP via langchain-mcp-adapters
StateConversation context + MEMORY.mdLangGraph checkpoints + UB
Cost$$$$ (premium)Cents (low-cost or free)
Best forArchitecture, key decisions, delegationBatch processing, research, classification

Mutual rescue

The dual-line architecture ensures resilience:

Captain's principle: "One goes down, the other can rescue." Neither line is disposable — both must be production-ready at all times.


Agent Anatomy — 13 Components

Line 1 designed the blueprint (Agent Anatomy 13 Components), and Line 2 implements it:

#ComponentDescription
1IdentityAgent name, role, ship assignment
2MemoryShort-term (context) + long-term (UB)
3PerceptionInput processing (text, images, files)
4ReasoningDecision-making logic
5PlanningTask decomposition and sequencing
6ActionTool execution capabilities
7LearningPattern recognition and correction capture
8CommunicationInter-agent messaging (mailbox)
9Self-monitoringHeartbeat, health checks
10GovernanceConstitution, cost awareness, HITL
11ContextSession management, handoff protocols
12ToolsMCP server connections
13StatePersistent state across sessions

Causal relationship: Line 1 blueprint → Line 2 implementation. Line 1 is the pioneer, not the laggard.


Line 2 ecosystem

The LangGraph line uses these verified packages:

PackageVersionPurpose
langgraph1.0.8Agent graph framework
langgraph-checkpoint-sqlite3.0.3State persistence
langchain-mcp-adapters0.2.1MCP tool bridge
langchain-google-genaiLatestGemini integration
langchain-groqLatestGroq integration

Choosing between lines

ScenarioRecommended line
Architecture design, key decisionsLine 1 (Claude)
Code editing, file operationsLine 1 (Claude)
Research and intel gatheringLine 2 (Groq/Gemini)
Content classificationLine 2 (MTAAA pipeline)
Batch data processingLine 2 (Groq = free)
Strategy analysisLine 1 (Claude Chat)
Emergency quota conservationLine 2 (all engines)

Non-dogmatic component sharing

Lines can cherry-pick components from each other without vendor lock-in:

  • Line 2 can use Claude as an engine for complex tasks (via dispatch_work_order with engine: "claude")
  • Line 1 can delegate to Groq/Gemini for cost optimization (via search_web, call_model)
  • Both share the same Cloud UB, WO system, and MCP infrastructure

The goal is agentic capability, not framework purity.


PageRelationship
Agent SDK QuickstartGet started with either line
Python SDK (LangGraph)Line 2 deep dive
TypeScript SDKLine 1 SDK reference
Line MigrationMoving between lines