Maigret (soxoj/maigret) is an open-source Python OSINT tool that searches a username across 3,000+ websites, social networks, and forums, then aggregates the results into a structured profile dossier. Designed for security researchers, investigative journalists, and red-team practitioners, it ships as a CLI and a Python library, with output formats including HTML, PDF, JSON, XMind, and a graph view that visualizes relationships across discovered accounts. The agent layer comes from extensible site definitions (over 3,000 maintained in YAML), automatic site classification, and an enrichment pipeline that pulls bio data, posts, and account metadata when available. Trended +1,117 stars on May 4, 2026, hitting 23,561 total — driven by the broader vertical-agent wave and growing OSINT tooling adoption among security teams. The name is a nod to Georges Simenon's fictional detective.
Maigret is a vertical OSINT agent: give it a username and it returns a structured dossier of where that handle exists across 3,000+ sites. The agent layer isn’t multi-LLM debate — it’s a focused search-and-enrichment pipeline with site classification, metadata extraction, and report generation. That focus is why it trends: when the job is “find every account belonging to this username,” generic LLM agents don’t beat a purpose-built tool with a curated site registry.
The repo hit 23,561 stars on May 4, 2026, with +1,117 added in the last 24 hours — part of the May 2026 vertical-agent surge documented in the AgentConn directory wave article. Security teams, investigative journalists, and red-team practitioners are the primary users, but the codebase has become a reference for “OSINT-as-an-agent” patterns: structured site registries, parallel scraping, and report synthesis.
Site coverage: 3,000+ sites maintained as YAML definitions in the repo. Categories include social networks, dating sites, forums, dev platforms (GitHub, GitLab, etc.), gaming, and country-specific services. Community PRs continuously expand and update the registry.
Output formats: HTML reports with clickable links, PDF exports for case files, JSON for downstream processing, XMind for analyst workflows, and a graph view that visualizes relationships across discovered accounts.
Enrichment: When a site exposes profile data (bio, post count, registration date, follower count), Maigret captures it. The structured output makes downstream analysis or LLM-based synthesis trivial.
Security research: Mapping attacker infrastructure or impersonation campaigns by username pivoting.
Investigative journalism: Tracking subjects across platforms while preserving evidence trails.
Red-team operations: Reconnaissance during authorized social-engineering engagements.
OSINT pipelines: Embedding Maigret as a Python library inside larger investigation workflows — feeding its structured output into downstream LLM agents for synthesis or report writing.
Maigret only surfaces public information. It does not bypass authentication, scrape behind paywalls, or extract private data. Users are responsible for compliance with site terms of service and applicable laws — this is an OSINT tool intended for legitimate research, not stalking or harassment.
An AI-powered academic search engine that finds and synthesizes evidence-based answers from peer-reviewed scientific research.
An AI research assistant that helps researchers search, analyze, and synthesize findings from academic papers at scale.
xAI's conversational AI with real-time X (Twitter) data access, web search, and image understanding capabilities.