Introduction

th is a single binary that reads Threads the way a search engine does. It asks threads.com for the server-rendered page behind a profile or post, parses the embedded JSON into typed records, and gets out of your way. There is no login, no browser, nothing to sign up for, and nothing to run alongside it.

How it is built

• A library package (threads) holds the crawler HTTP client, the server-rendered page parsers, the logged-out GraphQL queries, and the typed data models. It paces requests, sets an honest crawler User-Agent, and retries the transient failures any public site throws under load.
• A command tree (cli) wraps the library in subcommands with shared output formats and flags.
• A standalone pkg/thid classifies any handle, id, shortcode, or URL with no network and no other dependencies.
• One cmd/th entry point ties them together.

Scope

th is a read-only client over data Threads already serves publicly. It reads that data and shapes it for you. Private content stays private; when a target is walled, th exits 4 rather than returning nothing. That narrow scope keeps it a single small binary with no database, no daemon, and no setup. Optional token and session modes add depth on your own account, off by default.

Next: install it, then take the quick start.