Skip to content

ThomasMoreAI/legal-skills-open / br/litigation/skills/juriscan-brunobracaioli

juriscan-brunobracaioli

Verified

Análise forense de processos judiciais brasileiros. Extrai, classifica e cruza peças processuais, detecta contradições, calcula prazos CPC e gera relatórios de risco com exportação para Obsidian. Use quando mencionar: processo judicial, petição, sentença, acórdão, contradições, prazos, timeline, análise forense, Obsidian.

FileKindSize
.claude/agents/juriscan-advogado-autor.mdmarkdown4.3 KB
.claude/agents/juriscan-advogado-reu.mdmarkdown3.6 KB
.claude/agents/juriscan-auditor-processual.mdmarkdown7.0 KB
.claude/agents/juriscan-echo.mdmarkdown2.0 KB
.claude/agents/juriscan-parser.mdmarkdown4.7 KB
.claude/agents/juriscan-segmenter.mdmarkdown4.9 KB
.claude/agents/juriscan-sintetizador.mdmarkdown3.1 KB
.claude/agents/juriscan-verificador.mdmarkdown4.8 KB
.github/FUNDING.ymlconfig77 B
.github/workflows/ci.ymlconfig1.0 KB
.gitignoreother121 B
CHANGELOG.mdmarkdown20.6 KB
CLAUDE.mdmarkdown10.6 KB
LICENSEother1.0 KB
README.mdmarkdown16.9 KB
SKILL.mdskill md26.3 KB
assets/banner.svgother4.7 KB
docs/architecture.mdmarkdown10.3 KB
docs/index.htmlother4.3 KB
docs/operations.mdmarkdown6.7 KB
docs/subagent_contract.mdmarkdown6.0 KB
docs/subagents.mdmarkdown5.5 KB
install.shscript3.2 KB
pyproject.tomlmanifest272 B
references/agent_schemas/advogado_output.jsonconfig2.7 KB
references/agent_schemas/auditor_output.jsonconfig4.8 KB
references/agent_schemas/echo_output.jsonconfig666 B
references/agent_schemas/parser_output.jsonconfig4.4 KB
references/agent_schemas/recommendations_output.jsonconfig2.5 KB
references/agent_schemas/segmenter_output.jsonconfig2.9 KB
references/agent_schemas/sintetizador_output.jsonconfig2.7 KB
references/agent_schemas/verificador_output.jsonconfig2.3 KB
references/appellate_structure.mdmarkdown2.0 KB
references/binding_precedents.mdmarkdown2.2 KB
references/brazilian_legal_entities.mdmarkdown3.2 KB
references/chunk_analysis_schema.jsonconfig6.7 KB
references/cpc_prazos.jsonconfig4.4 KB
references/feriados_forenses.jsonconfig2.1 KB
references/output_schema.jsonconfig11.0 KB
references/output_schema_v2.jsonconfig11.0 KB
references/output_schema_v3.jsonconfig5.3 KB
references/piece_type_taxonomy.jsonconfig7.5 KB
references/prompt_templates.mdmarkdown11.8 KB
references/risk_scoring_rubric.mdmarkdown2.4 KB
references/whitelist_fontes.jsonconfig1.8 KB
requirements.txtmanifest250 B
scripts/__init__.pyscript0 B
scripts/agent_io.pyscript12.2 KB
scripts/analyzed_init.pyscript3.3 KB
scripts/audit.pyscript4.8 KB
scripts/cleanup_audit.pyscript2.0 KB
scripts/confidence_rules.pyscript6.4 KB
scripts/content_quality_check.pyscript11.1 KB
scripts/enrich_deterministic.pyscript10.3 KB
scripts/extract_and_chunk.pyscript14.3 KB
scripts/finalize.pyscript10.2 KB
scripts/finalize_legacy.pyscript7.4 KB
scripts/generate_report.pyscript29.8 KB
scripts/instance_tracker.pyscript6.7 KB
scripts/integrity_check.pyscript11.0 KB
scripts/integrity_gate.pyscript4.6 KB
scripts/legacy/__init__.pyscript0 B
scripts/legacy/contradiction_report.pyscript11.9 KB
scripts/legacy/risk_scorer.pyscript10.2 KB
scripts/merge_chunk_analysis.pyscript14.8 KB
scripts/migrate_v2_to_v3.pyscript3.1 KB
scripts/obsidian_export.pyscript28.0 KB
scripts/persist_chunks.pyscript6.2 KB
scripts/prazo_calculator.pyscript14.2 KB
scripts/report_metrics.pyscript8.4 KB
scripts/schema_validator.pyscript5.2 KB
scripts/utils/__init__.pyscript0 B
scripts/utils/cnj.pyscript4.4 KB
scripts/utils/dates.pyscript7.8 KB
scripts/utils/filenames.pyscript2.6 KB
scripts/utils/monetary.pyscript5.2 KB
tests/__init__.pyscript0 B
tests/conftest.pyscript1.1 KB
tests/fixtures/agent_io/echo_invalid_missing_field.jsonconfig45 B
tests/fixtures/agent_io/echo_invalid_wrong_const.jsonconfig70 B
tests/fixtures/agent_io/echo_valid.jsonconfig103 B
tests/fixtures/agent_io/segmenter_valid.jsonconfig359 B
tests/fixtures/sample_analyzed.jsonconfig10.0 KB
tests/fixtures/sample_analyzed_with_prazos.jsonconfig470 B
tests/fixtures/sample_chunk_acordao.txtmarkdown3.4 KB
tests/fixtures/sample_chunk_peticao.txtmarkdown3.4 KB
tests/fixtures/sample_chunk_sentenca.txtmarkdown2.9 KB
tests/fixtures/sample_index.jsonconfig1.6 KB
tests/golden/_pdf_builder.pyscript6.1 KB
tests/golden/processo_01_sintetico_simples/expected_invariants.jsonconfig993 B
tests/golden/processo_01_sintetico_simples/source.txtmarkdown3.3 KB
tests/golden/processo_02_sintetico_art942/expected_art_942.jsonconfig898 B
tests/golden/processo_02_sintetico_art942/expected_invariants.jsonconfig1.3 KB
tests/golden/processo_02_sintetico_art942/source.txtmarkdown7.8 KB
tests/test_agent_io.pyscript4.4 KB
tests/test_analyzed_init.pyscript5.6 KB
tests/test_audit.pyscript4.4 KB
tests/test_cnj.pyscript2.9 KB
tests/test_confidence_rules.pyscript4.8 KB
tests/test_content_quality_check.pyscript14.5 KB
tests/test_contradiction_report.pyscript3.5 KB
tests/test_dates.pyscript5.0 KB
tests/test_enrich_deterministic.pyscript6.3 KB
tests/test_filenames.pyscript2.1 KB
tests/test_finalize.pyscript5.8 KB
tests/test_finalize_legacy.pyscript6.5 KB
tests/test_generate_report.pyscript24.0 KB
tests/test_golden_regression.pyscript6.3 KB
tests/test_instance_tracker.pyscript4.2 KB
tests/test_integrity_check.pyscript4.8 KB
tests/test_integrity_gate.pyscript3.6 KB
tests/test_merge_chunk_analysis.pyscript20.9 KB
tests/test_migrate_v2_to_v3.pyscript4.3 KB
tests/test_monetary.pyscript3.4 KB
tests/test_obsidian_export_v3.pyscript6.9 KB
tests/test_persist_chunks.pyscript6.1 KB
tests/test_prazo_calculator.pyscript14.1 KB
tests/test_report_metrics.pyscript5.3 KB
tests/test_risk_scorer.pyscript5.7 KB
tests/test_whitelist_enforcement.pyscript6.1 KB

scripts/legacy/contradiction_report.py

View on GitHub
#!/usr/bin/env python3
"""
contradiction_report.py — Gera relatório de contradições entre peças processuais.

DEPRECATED (Phase 5 Step 5.1, 2026-04-11): moved to scripts/legacy/. Still
consumed by SKILL.md when `--pipeline=legacy` is explicit. The agents
pipeline uses juriscan-auditor-processual + juriscan-advogados instead.
Removal scheduled 3 releases after the Phase 6.4 default flip.

Compara as análises de todas as peças e identifica:
- Fatos divergentes (mesmo fato narrado diferentemente)
- Valores inconsistentes (números que mudam entre peças)
- Datas conflitantes (cronologia que não bate)
- Argumentos contraditórios (parte contradiz a si mesma)
- Jurisprudência conflitante (mesma tese, precedentes opostos)

Usage:
    python3 contradiction_report.py --analysis ./analysis/analyzed.json --output ./analysis/contradictions.json
"""

import warnings as _warnings

_warnings.warn(
    "scripts.legacy.contradiction_report is deprecated; prefer the agents pipeline.",
    DeprecationWarning,
    stacklevel=2,
)

import argparse
import json
import os
import re
import sys
from collections import defaultdict


def extract_monetary_values(text: str) -> list:
    """Extract monetary values from text."""
    patterns = [
        r'R\$\s*[\d.,]+',
        r'[\d.,]+\s*(?:reais|mil|milhão|milhões|bilhão|bilhões)',
    ]
    values = []
    for p in patterns:
        for m in re.finditer(p, text, re.IGNORECASE):
            values.append(m.group().strip())
    return values


def normalize_value(val_str: str) -> float | None:
    """Try to normalize a monetary string to float."""
    try:
        cleaned = re.sub(r'[R$\s]', '', val_str)
        cleaned = cleaned.replace('.', '').replace(',', '.')
        return float(cleaned)
    except:
        return None


# Phase 1 Step 1.3 — hierarchy awareness.
# Pieces organized by their position in the appellate chain. When the SAME
# monetary concept is valued differently across pieces that straddle these
# tiers, that is almost always a legitimate appellate reform, not a
# contradiction. We emit an INSTANCE_TRACKING note instead of a VALOR_INCONSISTENTE
# flag. True contradictions (e.g. sentença vs. sentença disagreeing) remain flagged.

_LOWER_INSTANCE_PIECES = {"SENTENÇA", "SENTENCA"}
_APPELLATE_PIECES = {
    "ACÓRDÃO",
    "ACORDAO",
    "DECISÃO MONOCRÁTICA",
    "DECISAO MONOCRATICA",
}


def _is_legitimate_reform(labels: list[str]) -> bool:
    """True when the set of labels spans the hierarchy (lower + appellate)."""
    normalized = {(l or "").upper() for l in labels}
    has_lower = bool(normalized & _LOWER_INSTANCE_PIECES)
    has_appellate = bool(normalized & _APPELLATE_PIECES)
    return has_lower and has_appellate


def find_value_inconsistencies(chunks: list) -> tuple[list, list]:
    """Find monetary contradictions. Returns (contradictions, instance_notes).

    instance_notes contains informational entries for legitimate appellate
    reforms (hierarchical value changes that are NOT contradictions).
    """
    contradictions: list = []
    instance_notes: list = []

    # Compare valor_causa across chunks
    causas = []
    for c in chunks:
        v = c.get('valores', {}).get('causa')
        if v:
            causas.append((c.get('label', '?'), v))

    if len(causas) > 1:
        normalized = [(label, val, normalize_value(val)) for label, val in causas]
        unique_vals = set(n for _, _, n in normalized if n is not None)
        if len(unique_vals) > 1:
            contradictions.append({
                'tipo': 'VALOR_INCONSISTENTE',
                'impacto': 'MÉDIO',
                'pecas': [label for label, _, _ in normalized],
                'descricao': f"Valor da causa diverge entre peças: {', '.join(f'{label}: {val}' for label, val, _ in normalized)}",
                'sugestao': 'Verificar se houve emenda à inicial alterando o valor da causa, ou se alguma peça cita valor incorreto.',
            })

    # Compare condenação across chunks — hierarchy-aware
    condenacoes = []
    for c in chunks:
        v = c.get('valores', {}).get('condenacao')
        if v:
            condenacoes.append((c.get('label', '?'), v))

    if len(condenacoes) > 1:
        normalized = [(label, val, normalize_value(val)) for label, val in condenacoes]
        unique_vals = set(n for _, _, n in normalized if n is not None)
        if len(unique_vals) > 1:
            labels_involved = [label for label, _, _ in normalized]
            if _is_legitimate_reform(labels_involved):
                # Appellate reform of a lower-court condenação. Not a contradiction.
                instance_notes.append({
                    'tipo': 'REFORMA_PARCIAL',
                    'impacto': 'INFO',
                    'pecas': labels_involved,
                    'descricao': (
                        "Valor de condenação alterado em grau recursal (reforma "
                        "legítima): "
                        + ', '.join(f'{label}: {val}' for label, val, _ in normalized)
                    ),
                    'sugestao': (
                        'Reforma de sentença por instância superior é comportamento '
                        'esperado do sistema recursal — não é contradição. Verificar '
                        'base de cálculo de honorários e juros sobre a nova base.'
                    ),
                })
            else:
                contradictions.append({
                    'tipo': 'VALOR_INCONSISTENTE',
                    'impacto': 'ALTO',
                    'pecas': labels_involved,
                    'descricao': f"Valor de condenação diverge: {', '.join(f'{label}: {val}' for label, val, _ in normalized)}",
                    'sugestao': 'Verificar embargos de declaração sobre erro material, ou se houve reforma parcial em instância superior.',
                })

    return contradictions, instance_notes


def find_date_conflicts(chunks: list) -> list:
    """Find chronological inconsistencies."""
    contradictions = []
    
    # Check if dates are in logical order
    dated_chunks = [(c.get('label', '?'), c.get('primary_date')) for c in chunks if c.get('primary_date')]
    
    # Check petition before contestação, etc.
    order_rules = [
        ('PETIÇÃO INICIAL', 'CONTESTAÇÃO', 'Contestação datada antes da Petição Inicial'),
        ('CONTESTAÇÃO', 'RÉPLICA', 'Réplica datada antes da Contestação'),
        ('SENTENÇA', 'APELAÇÃO', 'Apelação datada antes da Sentença'),
        ('APELAÇÃO', 'ACÓRDÃO', 'Acórdão datado antes da Apelação'),
    ]
    
    def parse_date(d):
        try:
            parts = d.split('/')
            if len(parts) == 3:
                return int(parts[2]) * 10000 + int(parts[1]) * 100 + int(parts[0])
        except:
            pass
        return 0
    
    chunk_dates = {c.get('label', ''): parse_date(c.get('primary_date', '')) 
                   for c in chunks if c.get('primary_date')}
    
    for before, after, desc in order_rules:
        d_before = chunk_dates.get(before, 0)
        d_after = chunk_dates.get(after, 0)
        if d_before and d_after and d_after < d_before:
            contradictions.append({
                'tipo': 'DATA_CONFLITANTE',
                'impacto': 'ALTO',
                'pecas': [before, after],
                'descricao': desc,
                'sugestao': 'Possível erro de digitalização/OCR nas datas, ou peça protocolada fora de ordem. Verificar nos autos originais.',
            })
    
    return contradictions


def find_fact_divergences(chunks: list) -> list:
    """Find facts narrated differently by different parties."""
    contradictions = []
    
    # Compare fatos_relevantes between autor-side and réu-side pieces
    autor_facts = []
    reu_facts = []
    
    autor_pieces = ['PETIÇÃO INICIAL', 'RÉPLICA', 'APELAÇÃO', 'MEMORIAIS']
    reu_pieces = ['CONTESTAÇÃO', 'CONTRARRAZÕES', 'RECONVENÇÃO']
    
    for c in chunks:
        label = c.get('label', '')
        facts = c.get('fatos_relevantes', [])
        if label in autor_pieces:
            autor_facts.extend([(label, f) for f in facts])
        elif label in reu_pieces:
            reu_facts.extend([(label, f) for f in facts])
    
    # Note: Deep semantic comparison would require LLM analysis.
    # This script flags potential divergences for human review.
    if autor_facts and reu_facts:
        contradictions.append({
            'tipo': 'FATO_DIVERGENTE',
            'impacto': 'MÉDIO',
            'pecas': list(set([l for l, _ in autor_facts] + [l for l, _ in reu_facts])),
            'descricao': f"Detectadas {len(autor_facts)} narração(ões) do autor e {len(reu_facts)} do réu. Comparação semântica requer análise com LLM — use o prompt de contradições do SKILL.md.",
            'sugestao': 'Execute a análise cruzada com Claude usando o prompt template de detecção de contradições para comparação semântica profunda.',
        })
    
    return contradictions


def find_jurisprudence_conflicts(chunks: list) -> list:
    """Find conflicting jurisprudence cited by opposing parties."""
    contradictions = []
    
    autor_pieces = {'PETIÇÃO INICIAL', 'RÉPLICA', 'APELAÇÃO', 'MEMORIAIS'}
    reu_pieces = {'CONTESTAÇÃO', 'CONTRARRAZÕES', 'RECONVENÇÃO'}
    
    autor_juris = set()
    reu_juris = set()
    
    for c in chunks:
        label = c.get('label', '')
        juris = c.get('jurisprudencia', [])
        if label in autor_pieces:
            autor_juris.update(juris)
        elif label in reu_pieces:
            reu_juris.update(juris)
    
    # Same precedent cited by both sides
    common = autor_juris & reu_juris
    if common:
        contradictions.append({
            'tipo': 'JURISPRUDÊNCIA_CONFLITANTE',
            'impacto': 'MÉDIO',
            'pecas': ['Polo Ativo', 'Polo Passivo'],
            'descricao': f"Mesma jurisprudência citada por ambas as partes: {', '.join(list(common)[:5])}",
            'sugestao': 'Verificar se as partes interpretam o mesmo precedente de formas diferentes. Oportunidade de argumentação.',
        })
    
    return contradictions


def generate_report(analysis: dict) -> dict:
    """Generate full contradiction report."""
    chunks = analysis.get('chunks', [])

    all_contradictions = []
    value_contradictions, instance_notes = find_value_inconsistencies(chunks)
    all_contradictions.extend(value_contradictions)
    all_contradictions.extend(find_date_conflicts(chunks))
    all_contradictions.extend(find_fact_divergences(chunks))
    all_contradictions.extend(find_jurisprudence_conflicts(chunks))

    # Sort by impact
    impact_order = {'ALTO': 0, 'MÉDIO': 1, 'BAIXO': 2}
    all_contradictions.sort(key=lambda x: impact_order.get(x.get('impacto', 'BAIXO'), 3))

    return {
        'total': len(all_contradictions),
        'by_type': {
            t: len([c for c in all_contradictions if c['tipo'] == t])
            for t in set(c['tipo'] for c in all_contradictions)
        },
        'by_impact': {
            i: len([c for c in all_contradictions if c['impacto'] == i])
            for i in ['ALTO', 'MÉDIO', 'BAIXO']
        },
        'contradictions': all_contradictions,
        'instance_tracking': instance_notes,
    }


def main():
    parser = argparse.ArgumentParser(description='Generate contradiction report')
    parser.add_argument('--analysis', '-a', required=True, help='Path to analyzed.json')
    parser.add_argument('--output', '-o', required=True, help='Output path for contradictions.json')
    args = parser.parse_args()
    
    with open(args.analysis, 'r', encoding='utf-8') as f:
        analysis = json.load(f)
    
    report = generate_report(analysis)
    
    with open(args.output, 'w', encoding='utf-8') as f:
        json.dump(report, f, ensure_ascii=False, indent=2)
    
    print(f"\n✅ Contradiction report generated!")
    print(f"   Total contradictions: {report['total']}")
    print(f"   By impact: {json.dumps(report['by_impact'])}")
    print(f"   By type: {json.dumps(report['by_type'])}")
    print(f"   Output: {args.output}")


if __name__ == '__main__':
    main()
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 196kUpdated Claude CodeCodexVerifiedInstall