geracao-perfis-psicologicos-e-roleplay

Extracts psychological character profiles from manuscripts and configures local LLMs for in-character interviews.

Updated Jul 19, 2026
One-click install
npx skills add https://github.com/Ryanzucchi/Eldritch_Lich --skill geracao-perfis-psicologicos-e-roleplay-ryanzucchi
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: geracao-perfis-psicologicos-e-roleplay
Source: https://github.com/Ryanzucchi/Eldritch_Lich/tree/main/.agents/skills/geracao-perfis-psicologicos-e-roleplay
Command: npx skills add https://github.com/Ryanzucchi/Eldritch_Lich --skill geracao-perfis-psicologicos-e-roleplay-ryanzucchi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Authors of fiction struggle to keep character behavior consistent across a long manuscript and lack a way to test dialogue and subtext interactively. This Skill derives structured psychological profiles from a character's actions in the narrative and powers a local chatbot that roleplays the character without breaking persona or inventing facts. ## Core Features & Use Cases - Automated Profile Extraction: Classifies character appearances as Agent or Patient in the narrative event graph and derives Big Five (OCEAN) personality traits from action patterns. - In-Character Chatbot (Roleplay): Builds a system prompt with biography, traits, and event history so a local LLM (e.g., Qwen-2.5-7B, Llama-3-8B) answers author interviews in first person as the character. - Consistency Auditing: Checks whether a scene being written respects the character's registered psychological profile, with graph-anchored context injection to prevent hallucinated facts. - Use Case: An author asks "Kael, why do you hate the sorcerer Érebo?" and the chatbot answers in first person using only facts anchored in the character's Temporal Entity Graph history. ## Quick Start Ask the assistant to generate a psychological profile for a character from the manuscript and start an in-character interview chat with that persona.

Frequently Asked Questions about geracao-perfis-psicologicos-e-roleplay

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate a psychological profile for a fictional character automatically?▼

The Skill scans the narrative event graph, classifies each character appearance as Agent or Patient, and derives Big Five personality traits from action proportions. It then compiles a Markdown character sheet with traits, social network, valence arc, and pending goals.

How to run an in-character chatbot with a local LLM?▼

Build a system prompt containing the character's biography, psychological traits, vocabulary style, and event history, with strict first-person roleplay instructions. A quantized local model such as Qwen-2.5-7B or Llama-3-8B is recommended for this task.

How do I prevent persona drift in roleplay language models?▼

Inject neighboring nodes from the character's Temporal Entity Graph into every conversation turn and add a rejection instruction for unanchored facts. Lowering the temperature (default 0.70) and capping context tokens (default 4096) further reduces hallucination.

What local LLM size works for character roleplay?▼

The Skill targets 7B-8B class quantized models like Qwen-2.5-7B or Llama-3-8B. The documentation notes that maintaining a stable persona over long conversations remains an open challenge for local 7B models, with a passive profile-generation fallback for v1.

When should I not use character roleplay generation?▼

Do not use it for factual wiki questions about the world, which belong to narrative knowledge retrieval, or for open brainstorming without defined characters. It also requires at least 10 historical events per character before automatic trait derivation triggers.