# cyberpradeep/elone_agent architecture

> Architecture diagram of the GitHub repository cyberpradeep/elone_agent, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-05). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/cyberpradeep/elone_agent
- Repository: https://github.com/cyberpradeep/elone_agent

## Overview

This repository implements a LangGraph support-ticket agent: it collects or clarifies ticket details, retrieves a matching SOP, plans and validates actions, executes steps through provider tools, checks outcomes, and produces a response or escalation. `main_graph.py` establishes intake as the entry point; agents then use state updates and dynamic routing. The map also shows the tool-provider layer and shared LLM configuration. Evidence is limited to the supplied excerpts and SOURCE INDEX; unsampled behavior and some routing details are not asserted.

Main entry points: [`agents/orchestration/node.py`](https://github.com/cyberpradeep/elone_agent/blob/main/agents/orchestration/node.py) (Orchestration agent), [`shared/llm_config.py`](https://github.com/cyberpradeep/elone_agent/blob/main/shared/llm_config.py) (LLM configuration), [`agents/retrieval/node.py`](https://github.com/cyberpradeep/elone_agent/blob/main/agents/retrieval/node.py) (SOP retrieval), [`shared/tool_catalog.py`](https://github.com/cyberpradeep/elone_agent/blob/main/shared/tool_catalog.py) (Tool catalog)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_workflow["Workflow control"]
  node_graph["Workflow graph<br/>[main_graph.py]"]
  node_state["Support state<br/>[state.py]"]
  node_orchestration["Orchestration agent<br/>[node.py]"]
  node_router["Workflow router<br/>[router.py]"]
end

subgraph group_ticket["Ticket processing"]
  node_intake["Ticket intake<br/>[node.py]"]
  node_retrieval["SOP retrieval<br/>[node.py]"]
  node_qdrant[("Qdrant index")]
end

subgraph group_actions["Plan and execution"]
  node_planning["Execution planning<br/>[node.py]"]
  node_validator["Plan validation<br/>[node.py]"]
  node_execution["Step execution<br/>[node.py]"]
  node_dispatch["Tool dispatch<br/>[dispatch.py]"]
  node_result["Outcome evaluation<br/>[node.py]"]
  node_escalation["Human escalation<br/>[node.py]"]
  node_summary["Resolution summary<br/>[node.py]"]
end

subgraph group_platform["Shared integrations"]
  node_catalog["Tool catalog<br/>[tool_catalog.py]"]
  node_mock["Mock provider<br/>[mock.py]"]
  node_auth0["Auth0 provider<br/>[auth0.py]"]
  node_systemops["System operations<br/>[system_ops.py]"]
  node_llm["LLM configuration<br/>[llm_config.py]"]
end

node_user(("Ticket submitter"))
node_llmservice(("LLM services"))
node_auth0service(("Auth0 service"))

node_user -->|"submits ticket"| node_graph
node_graph -->|"starts at"| node_intake
node_intake -->|"routes to"| node_retrieval
node_retrieval -->|"queries index"| node_qdrant
node_intake -->|"loads intents"| node_qdrant
node_retrieval -->|"routes to"| node_orchestration
node_orchestration -->|"selects route"| node_router
node_orchestration -->|"routes to"| node_retrieval
node_orchestration -->|"routes to"| node_planning
node_orchestration -->|"routes to"| node_validator
node_orchestration -->|"routes to"| node_execution
node_orchestration -->|"routes to"| node_escalation
node_planning -->|"returns plan"| node_orchestration
node_execution -->|"calls tool"| node_dispatch
node_dispatch -->|"uses catalog"| node_catalog
node_planning -->|"selects tools"| node_catalog
node_catalog -->|"registers tools"| node_mock
node_catalog -->|"registers tools"| node_auth0
node_catalog -->|"registers tools"| node_systemops
node_execution -->|"returns result"| node_orchestration
node_result -->|"returns decision"| node_orchestration
node_orchestration -.->|"routes to"| node_result
node_orchestration -.->|"routes to"| node_summary
node_orchestration -.->|"updates state"| node_state
node_intake -->|"gets model"| node_llm
node_retrieval -->|"gets model"| node_llm
node_planning -->|"gets model"| node_llm
node_result -->|"gets model"| node_llm
node_summary -->|"gets model"| node_llm
node_llm -->|"builds clients"| node_llmservice
node_auth0 -->|"calls API"| node_auth0service

click node_graph "https://github.com/cyberpradeep/elone_agent/blob/main/main_graph.py"
click node_state "https://github.com/cyberpradeep/elone_agent/blob/main/shared/state.py"
click node_orchestration "https://github.com/cyberpradeep/elone_agent/blob/main/agents/orchestration/node.py"
click node_router "https://github.com/cyberpradeep/elone_agent/blob/main/agents/orchestration/router.py"
click node_intake "https://github.com/cyberpradeep/elone_agent/blob/main/agents/intake/node.py"
click node_retrieval "https://github.com/cyberpradeep/elone_agent/blob/main/agents/retrieval/node.py"
click node_planning "https://github.com/cyberpradeep/elone_agent/blob/main/agents/execution_planning/node.py"
click node_validator "https://github.com/cyberpradeep/elone_agent/blob/main/agents/plan_validator/node.py"
click node_execution "https://github.com/cyberpradeep/elone_agent/blob/main/agents/execution/node.py"
click node_dispatch "https://github.com/cyberpradeep/elone_agent/blob/main/agents/execution/dispatch.py"
click node_result "https://github.com/cyberpradeep/elone_agent/blob/main/agents/execution_result/node.py"
click node_escalation "https://github.com/cyberpradeep/elone_agent/blob/main/agents/escalation/node.py"
click node_summary "https://github.com/cyberpradeep/elone_agent/blob/main/agents/summarization/node.py"
click node_catalog "https://github.com/cyberpradeep/elone_agent/blob/main/shared/tool_catalog.py"
click node_mock "https://github.com/cyberpradeep/elone_agent/blob/main/shared/providers/mock.py"
click node_auth0 "https://github.com/cyberpradeep/elone_agent/blob/main/shared/providers/auth0.py"
click node_systemops "https://github.com/cyberpradeep/elone_agent/blob/main/shared/providers/system_ops.py"
click node_llm "https://github.com/cyberpradeep/elone_agent/blob/main/shared/llm_config.py"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_graph,node_state,node_orchestration,node_router toneBlue
class node_intake,node_retrieval,node_qdrant toneAmber
class node_planning,node_validator,node_execution,node_dispatch,node_result,node_escalation,node_summary toneMint
class node_catalog,node_mock,node_auth0,node_systemops,node_llm toneRose
class node_user,node_llmservice,node_auth0service toneIndigo
```

## Components

### Workflow control

- [Workflow graph](https://github.com/cyberpradeep/elone_agent/blob/main/main_graph.py): `main_graph.py`
- [Support state](https://github.com/cyberpradeep/elone_agent/blob/main/shared/state.py): `shared/state.py`
- [Orchestration agent](https://github.com/cyberpradeep/elone_agent/blob/main/agents/orchestration/node.py): `agents/orchestration/node.py`
- [Workflow router](https://github.com/cyberpradeep/elone_agent/blob/main/agents/orchestration/router.py): `agents/orchestration/router.py`

### Ticket processing

- [Ticket intake](https://github.com/cyberpradeep/elone_agent/blob/main/agents/intake/node.py): `agents/intake/node.py`
- [SOP retrieval](https://github.com/cyberpradeep/elone_agent/blob/main/agents/retrieval/node.py): `agents/retrieval/node.py`
- **Qdrant index**

### Plan and execution

- [Execution planning](https://github.com/cyberpradeep/elone_agent/blob/main/agents/execution_planning/node.py): `agents/execution_planning/node.py`
- [Plan validation](https://github.com/cyberpradeep/elone_agent/blob/main/agents/plan_validator/node.py): `agents/plan_validator/node.py`
- [Step execution](https://github.com/cyberpradeep/elone_agent/blob/main/agents/execution/node.py): `agents/execution/node.py`
- [Tool dispatch](https://github.com/cyberpradeep/elone_agent/blob/main/agents/execution/dispatch.py): `agents/execution/dispatch.py`
- [Outcome evaluation](https://github.com/cyberpradeep/elone_agent/blob/main/agents/execution_result/node.py): `agents/execution_result/node.py`
- [Human escalation](https://github.com/cyberpradeep/elone_agent/blob/main/agents/escalation/node.py): `agents/escalation/node.py`
- [Resolution summary](https://github.com/cyberpradeep/elone_agent/blob/main/agents/summarization/node.py): `agents/summarization/node.py`

### Shared integrations

- [Tool catalog](https://github.com/cyberpradeep/elone_agent/blob/main/shared/tool_catalog.py): `shared/tool_catalog.py`
- [Mock provider](https://github.com/cyberpradeep/elone_agent/blob/main/shared/providers/mock.py): `shared/providers/mock.py`
- [Auth0 provider](https://github.com/cyberpradeep/elone_agent/blob/main/shared/providers/auth0.py): `shared/providers/auth0.py`
- [System operations](https://github.com/cyberpradeep/elone_agent/blob/main/shared/providers/system_ops.py): `shared/providers/system_ops.py`
- [LLM configuration](https://github.com/cyberpradeep/elone_agent/blob/main/shared/llm_config.py): `shared/llm_config.py`

### Other components

- **Ticket submitter**
- **LLM services**
- **Auth0 service**

## Connections

- Ticket submitter → Workflow graph: submits ticket
- Workflow graph → Ticket intake: starts at (evidence: `main_graph.py`)
- Ticket intake → SOP retrieval: routes to (evidence: `agents/intake/node.py`)
- SOP retrieval → Qdrant index: queries index (evidence: `agents/retrieval/node.py`)
- Ticket intake → Qdrant index: loads intents (evidence: `agents/intake/node.py`)
- SOP retrieval → Orchestration agent: routes to (evidence: `agents/retrieval/node.py`)
- Orchestration agent → Workflow router: selects route (evidence: `agents/orchestration/node.py`)
- Orchestration agent → SOP retrieval: routes to
- Orchestration agent → Execution planning: routes to
- Orchestration agent → Plan validation: routes to
- Orchestration agent → Step execution: routes to
- Orchestration agent → Human escalation: routes to
- Execution planning → Orchestration agent: returns plan (evidence: `agents/execution_planning/node.py`)
- Step execution → Tool dispatch: calls tool (evidence: `agents/execution/node.py`)
- Tool dispatch → Tool catalog: uses catalog (evidence: `agents/execution/dispatch.py`)
- Execution planning → Tool catalog: selects tools (evidence: `agents/execution_planning/node.py`)
- Tool catalog → Mock provider: registers tools (evidence: `shared/tool_catalog.py`)
- Tool catalog → Auth0 provider: registers tools (evidence: `shared/tool_catalog.py`)
- Tool catalog → System operations: registers tools (evidence: `shared/tool_catalog.py`)
- Step execution → Orchestration agent: returns result (evidence: `agents/execution/node.py`)
- Outcome evaluation → Orchestration agent: returns decision (evidence: `agents/execution_result/node.py`)
- Orchestration agent → Outcome evaluation: routes to
- Orchestration agent → Resolution summary: routes to
- Orchestration agent → Support state: updates state
- Ticket intake → LLM configuration: gets model (evidence: `agents/intake/node.py`)
- SOP retrieval → LLM configuration: gets model (evidence: `agents/retrieval/node.py`)
- Execution planning → LLM configuration: gets model (evidence: `agents/execution_planning/node.py`)
- Outcome evaluation → LLM configuration: gets model (evidence: `agents/execution_result/node.py`)
- Resolution summary → LLM configuration: gets model (evidence: `agents/summarization/node.py`)
- LLM configuration → LLM services: builds clients (evidence: `shared/llm_config.py`)
- Auth0 provider → Auth0 service: calls API (evidence: `shared/providers/auth0.py`)
