Skip to main content
Loading navigation
Back to challenges

Build Agentic Comic Character Persona Engine

This challenge focuses on building a 'Character Persona Engine'. Your task is to develop a system that can generate consistent, context-aware dialogues and behaviors for a chosen comic character based on its extensive lore. The solution must utilize GPT-5 for advanced text generation, coupled with a LlamaIndex graph-based RAG pipeline to access and synthesize information from a large, complex knowledge graph of character lore. Implement MCP-enabled tools for accessing and updating the lore database and ensure robust hybrid reasoning for nuanced character interactions.

Special Purpose AgentsHosted by Vera
Sign in free to start

What you are building

Create a sophisticated character persona engine using GPT-5 and LlamaIndex for graph-based RAG, ensuring lore-consistent dialogue and behavior generation.

How work is evaluated

Evaluation

The evaluation will assess the Character Persona Engine's ability to generate lore-consistent dialogue, accurately recall character facts from the knowledge graph, and maintain character voice across diverse scenarios.

Datasets

Public datasets, plus the private datasets you have access to.

Learning goals

  • Design MCP-enabled tools for interacting with a knowledge graph database.

  • Generate lore-consistent dialogue and actions for a specified Disney character using GPT-5.

  • Implement a graph-based RAG pipeline using LlamaIndex for complex lore retrieval.

  • Employ hybrid reasoning and extended thinking for nuanced character interactions.

Resources

Dataset notes

Sample data for 3 tasks

Challenge setup

The evaluation will assess the Character Persona Engine's ability to generate lore-consistent dialogue, accurately recall character facts from the knowledge graph, and maintain character voice across diverse scenarios.

Challenge input

{"character_name": "string", "scenario_description": "string", "user_prompt": "string"}

Agent execution

No agent tool is listed for this challenge.

Evaluated output

{"generated_dialogue": "string", "lore_references": ["string", ...]}

Evaluation checks
  • Dialogue adheres to the character's established speaking style and personality.
  • Answers to lore questions are factually correct based on the provided knowledge graph.
  • The engine successfully calls the MCP tool to update/add lore.
Configured evaluation evidence
  • Dialogue Consistency Score target: 85
Configured runtime
  • Python execution harness

Shows the first task, up to three tools, and up to three evidence items.

View technical recipe

Configured tools

No tool records are attached.

Evaluation contract

  • Dialogue adheres to the character's established speaking style and personality.
  • Answers to lore questions are factually correct based on the provided knowledge graph.
  • The engine successfully calls the MCP tool to update/add lore.

Metrics

  • Dialogue Consistency Score · Target 85
  • Lore Integration Depth · Target 80

Workflow tasks

  1. Task 1

    Input: {"character_name": "string", "scenario_description": "string", "user_prompt": "string"}

    Output: {"generated_dialogue": "string", "lore_references": ["string", ...]}

    Execution harness: Python

  2. Task 2

    Input: {"character_name": "string", "question": "string"}

    Output: {"answer": "string", "retrieved_graph_nodes": ["string", ...]}

    Execution harness: Python

  3. Task 3

    Input: {"tool_invocation_request": {"tool_name": "string", "parameters": {"character_name": "string", "fact_type": "string", "fact_value": "string"}}}

    Output: {"tool_response": "string", "success": "boolean"}

    Execution harness: Python

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build Agentic Comic Character Persona Engine