character-evidence-profiler

Profiles fictional characters from text evidence with confidence levels and source references.

Updated Aug 3, 2026
One-click install
npx skills add https://github.com/yeisme/auctra-novel-skills --skill character-evidence-profiler-yeisme
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: character-evidence-profiler
Source: https://github.com/yeisme/auctra-novel-skills/tree/main/character-modeling/character-evidence-profiler
Command: npx skills add https://github.com/yeisme/auctra-novel-skills --skill character-evidence-profiler-yeisme

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Writers and editors often guess at a character's personality, motivation, and voice without grounding those claims in what the character actually does or says on the page. This Skill builds character profiles strictly from observable evidence in drafts, scenes, dialogue, and dossiers, separating fact from inference so downstream writing decisions rest on verifiable material. ## Core Features & Use Cases - Evidence-First Profiling: Extracts actions, choices, wording, avoidance patterns, and relationship shifts, tagging each claim with a source reference and a high/medium/low confidence level. - Structured Fact Status: Distinguishes observed evidence, creative inference, author decisions, and open hypotheses, and classifies contradictory evidence as growth, disguise, perspective bias, or continuity error. - Use Case: Given three draft chapters of a novel, produce a character profile covering cognition, motivation, emotional patterns, relationships, and voice, with unknowns explicitly listed and each conclusion mapped to writable actions, dialogue, or choices. ## Quick Start Analyze the attached chapters and build an evidence-backed profile of the protagonist, separating observed facts from inference with confidence levels.

Frequently Asked Questions about character-evidence-profiler

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

FAQPage Schema
How do I analyze a fictional character's personality from a draft?▼

Extract observable evidence first: actions, choices, wording, avoidance, and stress reactions, each tagged with a source reference. Then select only the minimal facets that explain those choices, and label every conclusion as observed evidence, inference, author decision, or open hypothesis.

How to handle contradictory character behavior across chapters?▼

Classify each contradiction as a state change, relationship difference, disguise, growth, narrative perspective bias, or continuity error. This keeps genuine character complexity separate from drafting mistakes that need revision.

Can character profiling replace personality tests like MBTI?▼

No. The workflow explicitly forbids applying MBTI, archetypes, or psychological labels before examining text evidence. Labels may only be considered after observable behavior is documented, and a single test never substitutes for textual proof.

What are the limitations of evidence-based character analysis?▼

It cannot diagnose real people's mental illness, personality disorders, dangerousness, or deception from a few quotes. Missing evidence is recorded as unknown rather than treated as absence of a trait, and low-confidence claims state what material would raise confidence.

Does character evidence profiling work with copyrighted source material?▼

It prioritizes user-provided materials, short evidence excerpts, and summaries rather than large quotations from copyrighted works. Public-domain character material is an accepted input for analysis.