#!/usr/bin/env python3 """ OSINT Investigator β€” PDF Report Generator Converts a structured OSINT report (markdown or dict) into a clean PDF. Usage (from agent): python3 generate_pdf.py --title "Jack Gooding" --input report.md --output ~/Desktop/ Usage (direct): python3 generate_pdf.py --title "Target Name" --markdown "..." --output /path/to/dir """ import argparse import json import os import re import sys from datetime import datetime try: from fpdf import FPDF except ImportError: print("ERROR: fpdf2 not installed. Run: pip3 install fpdf2 --break-system-packages") sys.exit(1) # Confidence level styling CONFIDENCE = { "high": {"emoji": "🟒", "label": "HIGH", "r": 34, "g": 139, "b": 34}, "medium": {"emoji": "🟠", "label": "MED", "r": 255, "g": 140, "b": 0}, "low": {"emoji": "πŸ”΄", "label": "LOW", "r": 204, "g": 0, "b": 0}, "unverified": {"emoji": "βšͺ", "label": "UNVERIFIED", "r": 128, "g": 128, "b": 128}, } CONFIG_PATH = os.path.join(os.path.dirname(__file__), '..', 'config', 'osint_config.json') def load_config(): path = os.path.normpath(CONFIG_PATH) if os.path.exists(path): with open(path) as f: return json.load(f) return {"output": {"pdf_enabled": True, "pdf_include_sources": True}} def sanitize(text): """Replace unicode chars that fpdf latin-1 can't handle.""" replacements = { "\u2019": "'", "\u2018": "'", "\u201c": '"', "\u201d": '"', "\u2013": "-", "\u2014": "--", "\u2022": "*", "\u2026": "...", "\u00a0": " ", "\u2713": "βœ“", "\u2715": "x", # confidence emojis β†’ text "🟒": "[HIGH]", "🟠": "[MED]", "πŸ”΄": "[LOW]", "βšͺ": "[?]", "βœ…": "[OK]", "❌": "[NO]", "⚠️": "[WARN]", "πŸ”": "", "🏏": "", "πŸ’»": "", "πŸ“±": "", "🐦": "", "🏒": "", "πŸ“„": "", "πŸ”—": "", "🌐": "", "πŸ“ž": "", "πŸ“§": "", "πŸ—ΊοΈ": "", "πŸ–ΌοΈ": "", "πŸ“‹": "", "🎯": "", } for src, dst in replacements.items(): text = text.replace(src, dst) # Strip any remaining non-latin1 return text.encode('latin-1', errors='replace').decode('latin-1') class OSINTReport(FPDF): def __init__(self, title, target, date_str): super().__init__() self.report_title = title self.target = target self.date_str = date_str self.set_auto_page_break(auto=True, margin=15) def header(self): self.set_fill_color(20, 20, 40) self.rect(0, 0, 210, 18, 'F') self.set_font("Helvetica", "B", 11) self.set_text_color(255, 255, 255) self.set_xy(10, 4) self.cell(0, 10, sanitize(f"OSINT REPORT | {self.target} | {self.date_str}"), ln=False) self.set_text_color(0, 0, 0) self.ln(18) def footer(self): self.set_y(-12) self.set_font("Helvetica", "I", 8) self.set_text_color(150, 150, 150) self.cell(0, 10, f"Page {self.page_no()} | Generated by OSINT Investigator | CONFIDENTIAL", align="C") self.set_text_color(0, 0, 0) def cover_page(self, query, target_type, summary=None): self.add_page() # Dark header block self.set_fill_color(20, 20, 40) self.rect(0, 0, 210, 70, 'F') self.set_font("Helvetica", "B", 28) self.set_text_color(255, 255, 255) self.set_xy(15, 15) self.cell(0, 12, "OSINT REPORT", ln=True) self.set_font("Helvetica", "B", 18) self.set_text_color(100, 200, 255) self.set_x(15) self.cell(0, 10, sanitize(self.target), ln=True) self.set_font("Helvetica", "", 10) self.set_text_color(180, 180, 220) self.set_x(15) self.cell(0, 8, f"Generated: {self.date_str}", ln=True) self.set_text_color(0, 0, 0) self.set_y(80) # Meta boxes self.set_font("Helvetica", "B", 10) self.set_fill_color(240, 242, 248) self.set_x(15) self.cell(55, 8, "Target:", fill=True) self.set_font("Helvetica", "", 10) self.cell(0, 8, sanitize(self.target), ln=True) self.set_font("Helvetica", "B", 10) self.set_fill_color(240, 242, 248) self.set_x(15) self.cell(55, 8, "Target Type:", fill=True) self.set_font("Helvetica", "", 10) self.cell(0, 8, sanitize(target_type), ln=True) self.set_font("Helvetica", "B", 10) self.set_fill_color(240, 242, 248) self.set_x(15) self.cell(55, 8, "Query:", fill=True) self.set_font("Helvetica", "", 10) self.cell(0, 8, sanitize(query[:80]), ln=True) # Confidence legend self.ln(6) self.set_font("Helvetica", "B", 10) self.set_x(15) self.cell(0, 7, "Confidence Legend:", ln=True) legend = [ ("high", "Green [HIGH]", "Verified from multiple reliable sources"), ("medium", "Orange [MED]", "Likely correct, single source or unverified"), ("low", "Red [LOW]", "Possible match, little corroborating evidence"), ("unverified", "Grey [UNVERIFIED]", "User-provided context, not independently confirmed"), ] for conf_key, label, desc in legend: conf = CONFIDENCE[conf_key] self.set_font("Helvetica", "B", 9) self.set_text_color(conf['r'], conf['g'], conf['b']) self.set_x(20) self.cell(38, 6, label) self.set_font("Helvetica", "", 9) self.set_text_color(80, 80, 80) self.cell(0, 6, sanitize(desc), ln=True) self.set_text_color(0, 0, 0) if summary: self.ln(4) self.set_x(15) self.set_font("Helvetica", "B", 10) self.set_fill_color(230, 240, 255) self.cell(180, 7, "Executive Summary", fill=True, ln=True) self.set_font("Helvetica", "", 9) self.set_x(15) self.multi_cell(180, 5, sanitize(summary)) def section_header(self, title): self.ln(4) self.set_fill_color(20, 20, 40) self.set_text_color(255, 255, 255) self.set_font("Helvetica", "B", 11) self.set_x(10) self.cell(190, 8, sanitize(f" {title}"), fill=True, ln=True) self.set_text_color(0, 0, 0) self.ln(1) def finding_row(self, finding, source, confidence_key): """Render a single finding row with confidence badge.""" conf = CONFIDENCE.get(confidence_key.lower(), CONFIDENCE["low"]) y = self.get_y() # Confidence badge self.set_fill_color(conf['r'], conf['g'], conf['b']) self.set_text_color(255, 255, 255) self.set_font("Helvetica", "B", 7) self.set_x(10) self.cell(22, 6, conf['label'], fill=True, align="C") # Finding text self.set_text_color(0, 0, 0) self.set_font("Helvetica", "", 9) self.set_x(35) self.cell(105, 6, sanitize(finding[:90])) # Source self.set_font("Helvetica", "I", 7) self.set_text_color(100, 100, 100) self.cell(0, 6, sanitize(source[:45]), ln=True) self.set_text_color(0, 0, 0) # Light separator line self.set_draw_color(220, 220, 220) self.line(10, self.get_y(), 200, self.get_y()) def add_text_block(self, heading, body, confidence_key=None): """Add a named text block, optionally with a confidence badge.""" if heading: self.set_font("Helvetica", "B", 9) self.set_x(12) if confidence_key: conf = CONFIDENCE.get(confidence_key.lower(), CONFIDENCE["low"]) self.set_text_color(conf['r'], conf['g'], conf['b']) self.cell(25, 5, f"[{conf['label']}]") self.set_text_color(30, 30, 30) self.cell(0, 5, sanitize(heading), ln=True) else: self.set_text_color(30, 30, 30) self.cell(0, 5, sanitize(heading), ln=True) if body: self.set_font("Helvetica", "", 9) self.set_text_color(60, 60, 60) self.set_x(15) self.multi_cell(180, 4.5, sanitize(body)) self.set_text_color(0, 0, 0) self.ln(1) def sources_section(self, sources): self.section_header("Sources") self.set_font("Helvetica", "", 8) self.set_text_color(40, 40, 120) for i, src in enumerate(sources, 1): self.set_x(12) self.cell(8, 5, f"{i}.") self.multi_cell(178, 5, sanitize(src)) self.set_text_color(0, 0, 0) def parse_markdown_report(md_text): """Parse a markdown OSINT report into structured sections.""" sections = [] current_section = None current_body = [] sources = [] in_sources = False meta = {"target": "", "date": "", "query": "", "target_type": "", "summary": ""} lines = md_text.split('\n') for line in lines: # Extract meta from header lines if line.startswith("**Date:**"): meta["date"] = re.sub(r'\*\*Date:\*\*\s*', '', line).strip() elif line.startswith("**Target Type:**"): meta["target_type"] = re.sub(r'\*\*Target Type:\*\*\s*', '', line).strip() elif line.startswith("**Query:**"): meta["query"] = re.sub(r'\*\*Query:\*\*\s*', '', line).strip() # Top level H1 = report title / target elif line.startswith("# OSINT Report:"): meta["target"] = line.replace("# OSINT Report:", "").strip() # H2 sections elif line.startswith("## "): if current_section: sections.append((current_section, '\n'.join(current_body).strip())) current_section = line[3:].strip() current_body = [] in_sources = "Sources" in current_section # H3 subsections β€” treat as bold heading in body elif line.startswith("### "): current_body.append(f"\n{line[4:].strip()}:") # Source lines elif in_sources and line.strip().startswith("- http"): sources.append(line.strip()[2:]) elif in_sources and re.match(r'https?://', line.strip()): sources.append(line.strip()) else: current_body.append(line) if current_section: sections.append((current_section, '\n'.join(current_body).strip())) return meta, sections, sources def confidence_from_text(text): """Detect confidence level from text containing HIGH/MEDIUM/LOW/UNVERIFIED.""" text_upper = text.upper() if "HIGH" in text_upper: return "high" elif "MEDIUM" in text_upper or "MED" in text_upper: return "medium" elif "LOW" in text_upper: return "low" elif "UNVERIFIED" in text_upper: return "unverified" return None def generate_pdf(markdown_text, output_dir="~/Desktop", target_override=None): """Main entry point: convert markdown report to PDF.""" cfg = load_config() output_dir = os.path.expanduser(output_dir or cfg.get("output", {}).get("pdf_output_dir", "~/Desktop")) include_sources = cfg.get("output", {}).get("pdf_include_sources", True) meta, sections, sources = parse_markdown_report(markdown_text) target = target_override or meta.get("target", "Unknown Target") date_str = meta.get("date") or datetime.utcnow().strftime("%Y-%m-%d %H:%M UTC") pdf = OSINTReport( title=f"OSINT Report - {target}", target=target, date_str=date_str, ) # Cover page summary_text = "" for sec_name, sec_body in sections: if "summary" in sec_name.lower() or "identity" in sec_name.lower(): summary_text = sec_body[:400] break pdf.cover_page( query=meta.get("query", f"OSINT investigation: {target}"), target_type=meta.get("target_type", "Person"), summary=summary_text if summary_text else None, ) # Content pages pdf.add_page() for sec_name, sec_body in sections: if not sec_body.strip(): continue # Skip cover meta sections and sources (handled separately) if any(x in sec_name.lower() for x in ["sources", "confidence & gaps"]): continue pdf.section_header(sec_name) # Check if section has a markdown table (findings table) if "|" in sec_body and "---" in sec_body: # Parse table rows as finding rows rows = [r for r in sec_body.split('\n') if '|' in r and '---' not in r] # First row is header for row in rows[1:]: # skip header row cells = [c.strip() for c in row.split('|') if c.strip()] if len(cells) >= 3: finding = cells[0] source = cells[1] if len(cells) > 1 else "" conf_text = cells[2] if len(cells) > 2 else "" conf_key = confidence_from_text(conf_text) or "low" pdf.finding_row(finding, source, conf_key) else: # Render as text blocks, detect inline confidence markers paragraphs = re.split(r'\n{2,}', sec_body) for para in paragraphs: para = para.strip() if not para: continue conf_key = confidence_from_text(para) # Clean markdown formatting para_clean = re.sub(r'\*\*(.*?)\*\*', r'\1', para) para_clean = re.sub(r'\*(.*?)\*', r'\1', para_clean) para_clean = re.sub(r'`(.*?)`', r'\1', para_clean) para_clean = re.sub(r'\[([^\]]+)\]\([^\)]+\)', r'\1', para_clean) para_clean = re.sub(r'^[-*]\s+', '', para_clean, flags=re.MULTILINE) pdf.add_text_block(None, para_clean, confidence_key=conf_key) # Confidence summary section (if present) for sec_name, sec_body in sections: if "confidence" in sec_name.lower(): pdf.section_header("Confidence Summary") if "|" in sec_body and "---" in sec_body: rows = [r for r in sec_body.split('\n') if '|' in r and '---' not in r] for row in rows[1:]: cells = [c.strip() for c in row.split('|') if c.strip()] if len(cells) >= 3: finding = cells[0] source = cells[1] if len(cells) > 1 else "" conf_text = cells[2] if len(cells) > 2 else "" conf_key = confidence_from_text(conf_text) or "low" pdf.finding_row(finding, source, conf_key) break # Sources if include_sources and sources: pdf.add_page() pdf.sources_section(sources) # Save os.makedirs(output_dir, exist_ok=True) safe_target = re.sub(r'[^a-zA-Z0-9_\- ]', '', target).strip().replace(' ', '_') timestamp = datetime.utcnow().strftime("%Y%m%d_%H%M") filename = f"OSINT_{safe_target}_{timestamp}.pdf" filepath = os.path.join(output_dir, filename) pdf.output(filepath) return filepath if __name__ == "__main__": parser = argparse.ArgumentParser(description="Generate OSINT PDF report") parser.add_argument("--input", help="Path to markdown report file") parser.add_argument("--markdown", help="Markdown text directly") parser.add_argument("--target", help="Override target name") parser.add_argument("--output", default="~/Desktop", help="Output directory") args = parser.parse_args() md = "" if args.input: with open(args.input) as f: md = f.read() elif args.markdown: md = args.markdown else: print("Reading markdown from stdin...") md = sys.stdin.read() filepath = generate_pdf(md, output_dir=args.output, target_override=args.target) print(f"PDF saved: {filepath}")