ThomasMoreAI/legal-skills-open / br/criminal/skills/investigacao-policial-diegocamara89
investigacao-policial-diegocamara89
VerifiedSkill operacional para agentes de policia investigadores focada em producao de provas, estabelecimento de autoria e materialidade, e acompanhamento de diligencias. Use esta skill sempre que o usuario mencionar "o que devo coletar", "como provar autoria", "quais provas tenho", "o que esta pendente", "montar tabela de IPs", "cruzar IPs", "expandir suspeito", "expandir rede", "relatorio de diligencias", "cadeia de custodia", "como requisitar dados ao WhatsApp", "status da investigacao", "materialidade", "autoria", "laranja", "conta laranja", "IMEI", "cruzamento de fontes", "convergencia", "golpe do Pix", "golpe do falso advogado", "estelionato digital", "rastrear autor", "o que falta no IP", "pendencias da investigacao", "processar documentos em lote", "quem fez isso", "de onde veio o dinheiro", ou quando /investigacao for invocado. Aplicavel a qualquer tipo de fraude (estelionato, falso advogado, fraude eletronica, Pix, conta laranja, organizacao criminosa). NAO use esta skill para tipificacao penal, analise de prescricao ou fundamentacao juridica — essas ficam com a skill relatorio-final-ip. NAO use para catalogar quais diligencias existem por tipo penal — isso fica com a skill diligencias-iniciais. Mesmo que o usuario nao use essas palavras exatas, ative esta skill sempre que o contexto envolver acompanhamento ativo de uma investigacao de fraude em andamento.
| File | Kind | Size |
|---|---|---|
| LICENSE | other | 1.1 KB |
| SKILL.md | skill md | 29.3 KB |
| agents/extrator.md | markdown | 7.3 KB |
| agents/openai.yaml | config | 383 B |
| agents/validador.md | markdown | 8.2 KB |
| assets/template_pcrn_drcc.docx | binary | 91.1 KB |
| references/analise_vinculos.md | markdown | 7.0 KB |
| references/cadeia_custodia.md | markdown | 4.1 KB |
| references/categorias_prova.md | markdown | 7.8 KB |
| references/estabelecimento_autoria.md | markdown | 7.1 KB |
| references/expansao_rede.md | markdown | 6.7 KB |
| references/metodologia_ip.md | markdown | 8.5 KB |
| references/orquestracao.md | markdown | 14.1 KB |
| references/requisitos_judiciais.md | markdown | 4.2 KB |
| references/schemas_multiagente.md | markdown | 9.6 KB |
| scripts/auditoria.py | script | 10.0 KB |
| scripts/extrator_verificacao.py | script | 17.1 KB |
| scripts/gerar_docx_relatorio.py | script | 22.6 KB |
| scripts/pre_processador.py | script | 11.0 KB |
| templates/checklist_provas_fraude.md | markdown | 3.1 KB |
| templates/relatorio_diligencias.md | markdown | 2.9 KB |
| templates/relatorio_missao_policial.md | markdown | 12.0 KB |
| templates/tabela_ip_cruzamento.md | markdown | 2.6 KB |
scripts/gerar_docx_relatorio.py
View on GitHub"""Generate a .docx report from a Markdown source using the PCRN/DRCC template.
Features:
- Official state header with police badge and state coat of arms
- SEGREDO DE JUSTICA box (optional)
- Identification box with IP, investigados, tipificacao (optional)
- Section headings (##) rendered as gray bordered boxes
- GFM tables rendered as proper Word tables with header shading
- Mermaid code blocks rendered as PNG images via mmdc (if available)
- Times New Roman 12pt, ABNT margins, 1.5 line spacing
- Footer with procedure reference and page numbers
"""
from __future__ import annotations
import argparse
import os
import re
import subprocess
import sys
import tempfile
from pathlib import Path
from docx import Document
from docx.enum.text import WD_ALIGN_PARAGRAPH
from docx.oxml import OxmlElement
from docx.oxml.ns import qn
from docx.shared import Cm, Pt, RGBColor, Twips
# ── Constants ──────────────────────────────────────────────────────────────
FONT_NAME = "Times New Roman"
FONT_SIZE = Pt(12)
FONT_SIZE_TABLE = Pt(10)
FONT_SIZE_HALF_POINTS = "24" # 12pt in half-points
LINE_SPACING = 1.5
# ── Helpers ────────────────────────────────────────────────────────────────
def _set_cell_borders(cell, val="single", sz="6", color="000000"):
"""Set all four borders on a table cell."""
tc = cell._tc
tcPr = tc.find(qn("w:tcPr"))
if tcPr is None:
tcPr = OxmlElement("w:tcPr")
tc.insert(0, tcPr)
tcBorders = tcPr.find(qn("w:tcBorders"))
if tcBorders is None:
tcBorders = OxmlElement("w:tcBorders")
tcPr.append(tcBorders)
for edge in ("top", "left", "bottom", "right"):
el = OxmlElement(f"w:{edge}")
el.set(qn("w:val"), val)
el.set(qn("w:sz"), sz)
el.set(qn("w:color"), color)
el.set(qn("w:space"), "0")
existing = tcBorders.find(qn(f"w:{edge}"))
if existing is not None:
tcBorders.remove(existing)
tcBorders.append(el)
def _set_cell_shading(cell, fill_color: str):
"""Set background shading on a table cell."""
tc = cell._tc
tcPr = tc.find(qn("w:tcPr"))
if tcPr is None:
tcPr = OxmlElement("w:tcPr")
tc.insert(0, tcPr)
shd = OxmlElement("w:shd")
shd.set(qn("w:val"), "clear")
shd.set(qn("w:fill"), fill_color)
tcPr.append(shd)
def _set_table_no_borders(table):
"""Remove all table-level borders (borders set at cell level instead)."""
tbl = table._tbl
tblPr = tbl.find(qn("w:tblPr"))
if tblPr is None:
tblPr = OxmlElement("w:tblPr")
tbl.insert(0, tblPr)
tblBorders = OxmlElement("w:tblBorders")
for edge in ("top", "left", "bottom", "right", "insideH", "insideV"):
el = OxmlElement(f"w:{edge}")
el.set(qn("w:val"), "nil")
el.set(qn("w:sz"), "0")
el.set(qn("w:color"), "000000")
el.set(qn("w:space"), "0")
tblBorders.append(el)
existing = tblPr.find(qn("w:tblBorders"))
if existing is not None:
tblPr.remove(existing)
tblPr.append(tblBorders)
def _set_table_width(table, width_cm: float):
"""Set preferred table width in cm."""
tbl = table._tbl
tblPr = tbl.find(qn("w:tblPr"))
if tblPr is None:
tblPr = OxmlElement("w:tblPr")
tbl.insert(0, tblPr)
tblW = OxmlElement("w:tblW")
twips_val = int(width_cm * 567)
tblW.set(qn("w:w"), str(twips_val))
tblW.set(qn("w:type"), "dxa")
existing = tblPr.find(qn("w:tblW"))
if existing is not None:
tblPr.remove(existing)
tblPr.append(tblW)
def _make_run(text: str, bold=False, font_size=None, font_name=None):
"""Create a run element with specified formatting."""
r = OxmlElement("w:r")
rPr = OxmlElement("w:rPr")
if bold:
b = OxmlElement("w:b")
b.set(qn("w:val"), "1")
rPr.append(b)
if font_name:
rFonts = OxmlElement("w:rFonts")
rFonts.set(qn("w:ascii"), font_name)
rFonts.set(qn("w:hAnsi"), font_name)
rPr.append(rFonts)
if font_size:
sz = OxmlElement("w:sz")
sz.set(qn("w:val"), font_size)
rPr.append(sz)
szCs = OxmlElement("w:szCs")
szCs.set(qn("w:val"), font_size)
rPr.append(szCs)
r.append(rPr)
t = OxmlElement("w:t")
t.set(qn("xml:space"), "preserve")
t.text = text
r.append(t)
return r
# ── Document configuration ────────────────────────────────────────────────
def configure_styles(document: Document):
"""Set up the Normal style for the document."""
normal = document.styles["Normal"]
normal.font.name = FONT_NAME
rPr = normal._element.find(qn("w:rPr"))
if rPr is None:
rPr = OxmlElement("w:rPr")
normal._element.append(rPr)
rFonts = rPr.find(qn("w:rFonts"))
if rFonts is None:
rFonts = OxmlElement("w:rFonts")
rPr.append(rFonts)
rFonts.set(qn("w:ascii"), FONT_NAME)
rFonts.set(qn("w:hAnsi"), FONT_NAME)
rFonts.set(qn("w:eastAsia"), FONT_NAME)
normal.font.size = FONT_SIZE
pf = normal.paragraph_format
pf.line_spacing = LINE_SPACING
pf.space_after = Pt(0)
pf.space_before = Pt(0)
def configure_footer(section, footer_text: str):
"""Add footer with procedure reference and page number."""
footer = section.footer
footer.is_linked_to_previous = False
paragraph = footer.paragraphs[0]
paragraph.alignment = WD_ALIGN_PARAGRAPH.CENTER
if footer_text:
run = paragraph.add_run(f"{footer_text} | P\u00e1gina ")
run.font.name = FONT_NAME
run.font.size = Pt(9)
fld = OxmlElement("w:fldSimple")
fld.set(qn("w:instr"), "PAGE")
r = OxmlElement("w:r")
rPr = OxmlElement("w:rPr")
sz = OxmlElement("w:sz")
sz.set(qn("w:val"), "18")
rPr.append(sz)
rFonts = OxmlElement("w:rFonts")
rFonts.set(qn("w:ascii"), FONT_NAME)
rFonts.set(qn("w:hAnsi"), FONT_NAME)
rPr.append(rFonts)
r.append(rPr)
fld.append(r)
paragraph._p.append(fld)
# ── Preamble elements ─────────────────────────────────────────────────────
def add_segredo_box(document: Document):
"""Add the 'SEGREDO DE JUSTICA' box: gray background, thick border."""
table = document.add_table(rows=1, cols=1)
_set_table_no_borders(table)
cell = table.cell(0, 0)
_set_cell_borders(cell, val="single", sz="12", color="000000")
_set_cell_shading(cell, "d9d9d9")
p = cell.paragraphs[0]
p.alignment = WD_ALIGN_PARAGRAPH.CENTER
run = p.add_run("SEGREDO DE JUSTI\u00c7A")
run.bold = True
run.font.name = FONT_NAME
run.font.size = FONT_SIZE
document.add_paragraph() # spacing
def add_identification_box(
document: Document,
ip_numero: str,
investigados: str | None = None,
tipificacao: list[str] | None = None,
):
"""Add the identification box with IP, investigados, tipificacao."""
table = document.add_table(rows=1, cols=1)
_set_table_no_borders(table)
cell = table.cell(0, 0)
_set_cell_borders(cell, val="single", sz="6", color="000000")
# IP line
p_ip = cell.paragraphs[0]
p_ip.alignment = WD_ALIGN_PARAGRAPH.JUSTIFY
run_label = p_ip.add_run("Inqu\u00e9rito Policial: ")
run_label.bold = True
run_label.font.name = FONT_NAME
run_label.font.size = FONT_SIZE
run_val = p_ip.add_run(ip_numero)
run_val.font.name = FONT_NAME
run_val.font.size = FONT_SIZE
# Investigados line
if investigados:
p_inv = cell.add_paragraph()
p_inv.alignment = WD_ALIGN_PARAGRAPH.JUSTIFY
run_label = p_inv.add_run("Investigados: ")
run_label.bold = True
run_label.font.name = FONT_NAME
run_label.font.size = FONT_SIZE
run_val = p_inv.add_run(investigados)
run_val.font.name = FONT_NAME
run_val.font.size = FONT_SIZE
# Tipificacao lines
if tipificacao:
p_tip = cell.add_paragraph()
p_tip.alignment = WD_ALIGN_PARAGRAPH.JUSTIFY
run_label = p_tip.add_run("Tipifica\u00e7\u00e3o Provis\u00f3ria:")
run_label.bold = True
run_label.font.name = FONT_NAME
run_label.font.size = FONT_SIZE
for tipo in tipificacao:
p_t = cell.add_paragraph()
run_t = p_t.add_run(tipo)
run_t.font.name = FONT_NAME
run_t.font.size = FONT_SIZE
document.add_paragraph() # spacing
def add_section_header_box(document: Document, title: str):
"""Add a section header in a gray bordered box."""
table = document.add_table(rows=1, cols=1)
_set_table_no_borders(table)
cell = table.cell(0, 0)
_set_cell_borders(cell, val="single", sz="12", color="000000")
_set_cell_shading(cell, "d9d9d9")
p = cell.paragraphs[0]
p.alignment = WD_ALIGN_PARAGRAPH.JUSTIFY
run = p.add_run(title)
run.bold = True
run.font.name = FONT_NAME
run.font.size = FONT_SIZE
# ── GFM Table handling ────────────────────────────────────────────────────
def _is_table_separator(line: str) -> bool:
"""Check if a line is a GFM table separator row (|---|---|)."""
stripped = line.strip()
return bool(re.match(r"^\|?\s*[-:]+[-| :]*$", stripped))
def _parse_table_row(line: str) -> list[str]:
"""Parse a GFM table row into a list of cell strings."""
stripped = line.strip()
if stripped.startswith("|"):
stripped = stripped[1:]
if stripped.endswith("|"):
stripped = stripped[:-1]
return [c.strip() for c in stripped.split("|")]
def add_md_table(document: Document, table_lines: list[str]):
"""Render a GFM markdown table as a properly formatted Word table."""
if len(table_lines) < 2:
return
header = _parse_table_row(table_lines[0])
# table_lines[1] is the separator row — skip
data_rows = [_parse_table_row(ln) for ln in table_lines[2:]]
n_cols = len(header)
if n_cols == 0:
return
# Normalize all rows to same column count
for row in data_rows:
while len(row) < n_cols:
row.append("")
total_rows = 1 + len(data_rows)
table = document.add_table(rows=total_rows, cols=n_cols)
_set_table_no_borders(table)
_set_table_width(table, 16.0)
# Distribute columns evenly across 16cm
col_width = Cm(16.0 / n_cols)
for col in table.columns:
col.width = col_width
# Header row — gray background, bold, centered
for j, cell_text in enumerate(header[:n_cols]):
cell = table.cell(0, j)
_set_cell_borders(cell, val="single", sz="6", color="000000")
_set_cell_shading(cell, "d9d9d9")
p = cell.paragraphs[0]
p.alignment = WD_ALIGN_PARAGRAPH.CENTER
p.paragraph_format.line_spacing = LINE_SPACING
run = p.add_run(cell_text)
run.bold = True
run.font.name = FONT_NAME
run.font.size = FONT_SIZE_TABLE
# Data rows
for i, row in enumerate(data_rows):
for j in range(n_cols):
cell_text = row[j] if j < len(row) else ""
cell = table.cell(i + 1, j)
_set_cell_borders(cell, val="single", sz="6", color="000000")
p = cell.paragraphs[0]
p.alignment = WD_ALIGN_PARAGRAPH.JUSTIFY
p.paragraph_format.line_spacing = LINE_SPACING
add_inline_formatting(p, cell_text, font_size=FONT_SIZE_TABLE)
document.add_paragraph() # spacing after table
# ── Mermaid rendering ─────────────────────────────────────────────────────
def _find_mmdc() -> str | None:
"""Locate the mmdc executable (supports Windows .cmd wrapper)."""
import shutil
for name in ("mmdc", "mmdc.cmd"):
found = shutil.which(name)
if found:
return found
# Common Windows npm global bin locations
for candidate in [
os.path.expanduser("~\\AppData\\Roaming\\npm\\mmdc.cmd"),
os.path.expanduser("~\\AppData\\Roaming\\npm\\mmdc"),
]:
if os.path.isfile(candidate):
return candidate
return None
def render_mermaid_png(mermaid_code: str, idx: int, temp_dir: str) -> str | None:
"""Render mermaid diagram code to PNG via mmdc. Returns PNG path or None."""
mmdc = _find_mmdc()
if not mmdc:
print("WARNING: mmdc not found. Install with: npm install -g @mermaid-js/mermaid-cli", file=sys.stderr)
return None
input_file = os.path.join(temp_dir, f"mermaid_{idx}.mmd")
output_file = os.path.join(temp_dir, f"mermaid_{idx}.png")
try:
with open(input_file, "w", encoding="utf-8") as f:
f.write(mermaid_code)
result = subprocess.run(
[mmdc, "-i", input_file, "-o", output_file, "-b", "white", "--width", "1400"],
capture_output=True,
text=True,
timeout=90,
shell=(os.name == "nt"),
)
if result.returncode == 0 and os.path.exists(output_file):
return output_file
print(f"mmdc error (diagram {idx}): {result.stderr[:300]}", file=sys.stderr)
return None
except Exception as e:
print(f"Mermaid render failed (diagram {idx}): {e}", file=sys.stderr)
return None
# ── Markdown parsing ──────────────────────────────────────────────────────
def add_inline_formatting(paragraph, text: str, font_size=None):
"""Parse **bold** and [A PREENCHER] markers and add formatted runs."""
if font_size is None:
font_size = FONT_SIZE
parts = re.split(r"(\*\*.*?\*\*)", text)
for part in parts:
if not part:
continue
is_bold = part.startswith("**") and part.endswith("**")
content = part[2:-2] if is_bold else part
if is_bold:
run = paragraph.add_run(content)
run.bold = True
run.font.name = FONT_NAME
run.font.size = font_size
else:
# Split by [A PREENCHER] to apply red color
subparts = re.split(r"(\[A PREENCHER\])", content)
for subpart in subparts:
if not subpart:
continue
run = paragraph.add_run(subpart)
run.font.name = FONT_NAME
run.font.size = font_size
if subpart == "[A PREENCHER]":
run.font.color.rgb = RGBColor(0xFF, 0x00, 0x00)
def parse_markdown(
document: Document,
text: str,
skip_ip_line: bool = False,
temp_dir: str | None = None,
):
"""Parse Markdown text and add formatted content to the document."""
lines = text.splitlines()
mermaid_counter = [0]
i = 0
while i < len(lines):
line = lines[i].rstrip()
i += 1
if not line.strip():
document.add_paragraph()
continue
# Skip IP reference line if already rendered in identification box
if skip_ip_line and re.match(
r"^\s*Inqu[e\u00e9]rito Policial\s+n\.\s*\d", line, re.IGNORECASE
):
continue
# H1 title
if line.startswith("# "):
p = document.add_paragraph()
p.alignment = WD_ALIGN_PARAGRAPH.CENTER
p.paragraph_format.space_after = Pt(12)
run = p.add_run(line[2:].strip().upper())
run.bold = True
run.font.name = FONT_NAME
run.font.size = Pt(14)
continue
# H2 section -> gray bordered box
if line.startswith("## "):
add_section_header_box(document, line[3:].strip())
document.add_paragraph() # spacing after box
continue
# H3 subsection
if line.startswith("### "):
p = document.add_paragraph()
p.alignment = WD_ALIGN_PARAGRAPH.JUSTIFY
p.paragraph_format.space_before = Pt(6)
run = p.add_run(line[4:].strip())
run.bold = True
run.font.name = FONT_NAME
run.font.size = FONT_SIZE
continue
# Numbered list
if re.match(r"^\d+\.\s+", line):
p = document.add_paragraph(style="List Number")
p.alignment = WD_ALIGN_PARAGRAPH.JUSTIFY
p.paragraph_format.line_spacing = LINE_SPACING
add_inline_formatting(p, re.sub(r"^\d+\.\s+", "", line))
continue
# Bullet list
if line.startswith("- "):
p = document.add_paragraph(style="List Bullet")
p.alignment = WD_ALIGN_PARAGRAPH.JUSTIFY
p.paragraph_format.line_spacing = LINE_SPACING
add_inline_formatting(p, line[2:].strip())
continue
# GFM table: line starts with | and next line is a separator row
if line.strip().startswith("|") and i < len(lines) and _is_table_separator(lines[i]):
table_lines = [line]
while i < len(lines) and lines[i].rstrip().strip().startswith("|"):
table_lines.append(lines[i].rstrip())
i += 1
add_md_table(document, table_lines)
continue
# Mermaid fenced code block
if line.strip() == "```mermaid":
mermaid_lines = []
while i < len(lines) and lines[i].rstrip().strip() != "```":
mermaid_lines.append(lines[i])
i += 1
if i < len(lines):
i += 1 # consume closing ```
mermaid_code = "\n".join(mermaid_lines)
mermaid_counter[0] += 1
png_path = render_mermaid_png(mermaid_code, mermaid_counter[0], temp_dir) if temp_dir else None
if png_path:
document.add_picture(png_path, width=Cm(15.5))
document.paragraphs[-1].alignment = WD_ALIGN_PARAGRAPH.CENTER
else:
# Fallback: italic note
p = document.add_paragraph()
p.alignment = WD_ALIGN_PARAGRAPH.CENTER
run = p.add_run(f"[Diagrama {mermaid_counter[0]} — disponível no arquivo mermaid_diagramas.md]")
run.italic = True
run.font.name = FONT_NAME
run.font.size = FONT_SIZE
document.add_paragraph()
continue
# Any other fenced code block — skip content
if line.strip().startswith("```"):
while i < len(lines) and not lines[i].rstrip().strip().startswith("```"):
i += 1
if i < len(lines):
i += 1
continue
# Horizontal rule — skip
if re.match(r"^-{3,}$", line.strip()):
continue
# Regular paragraph
p = document.add_paragraph()
p.alignment = WD_ALIGN_PARAGRAPH.JUSTIFY
p.paragraph_format.first_line_indent = Pt(28)
p.paragraph_format.line_spacing = LINE_SPACING
add_inline_formatting(p, line)
# ── Main ───────────────────────────────────────────────────────────────────
def find_template(script_dir: Path) -> Path | None:
"""Look for the template in standard locations."""
candidates = [
script_dir / ".." / "assets" / "template_pcrn_drcc.docx",
script_dir / "assets" / "template_pcrn_drcc.docx",
Path.cwd() / "assets" / "template_pcrn_drcc.docx",
]
for c in candidates:
if c.resolve().exists():
return c.resolve()
return None
def main():
parser = argparse.ArgumentParser(
description="Generate a .docx report from Markdown using the PCRN/DRCC template."
)
parser.add_argument("--input", required=True, help="Path to Markdown source file")
parser.add_argument("--output", required=True, help="Path to output .docx file")
parser.add_argument("--template", default=None, help="Path to template .docx (auto-detected if omitted)")
parser.add_argument("--segredo-justica", action="store_true", help="Add SEGREDO DE JUSTICA box at the top")
parser.add_argument("--ip-numero", default=None, help="IP number for identification box")
parser.add_argument("--investigados", default=None, help="Investigados string for identification box")
parser.add_argument("--tipificacao", action="append", default=[], help="Tipificacao line (repeatable)")
parser.add_argument("--footer-text", default=None, help="Footer text (defaults to IP reference)")
# Backward compatibility
parser.add_argument("--header-line", action="append", default=[])
args = parser.parse_args()
source = Path(args.input)
target = Path(args.output)
target.parent.mkdir(parents=True, exist_ok=True)
# Locate template
template_path = args.template
if template_path is None:
template_path = find_template(Path(__file__).parent)
if template_path and Path(template_path).exists():
document = Document(str(template_path))
print(f"Using template: {template_path}", file=sys.stderr)
else:
document = Document()
print("WARNING: Template not found — creating blank document.", file=sys.stderr)
section = document.sections[0]
section.page_width = Cm(21.0)
section.page_height = Cm(29.7)
sectPr = section._sectPr
pgMar = sectPr.find(qn("w:pgMar"))
if pgMar is not None:
pgMar.set(qn("w:top"), "1701")
pgMar.set(qn("w:bottom"), "1134")
pgMar.set(qn("w:left"), "1701")
pgMar.set(qn("w:right"), "1134")
configure_styles(document)
footer_text = args.footer_text
if footer_text is None and args.ip_numero:
footer_text = f"IP n. {args.ip_numero}"
configure_footer(document.sections[0], footer_text or "")
if args.segredo_justica:
add_segredo_box(document)
if args.ip_numero:
add_identification_box(
document,
ip_numero=args.ip_numero,
investigados=args.investigados,
tipificacao=args.tipificacao if args.tipificacao else None,
)
# Use a temp directory for Mermaid PNG files
with tempfile.TemporaryDirectory() as temp_dir:
md_text = source.read_text(encoding="utf-8")
parse_markdown(document, md_text, skip_ip_line=bool(args.ip_numero), temp_dir=temp_dir)
document.save(str(target))
print(target.resolve())
if __name__ == "__main__":
main()
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |