What it does

Emulo mines the messages you typed to your coding agents into a private profile called you.md. It reads session logs from Claude Code, Codex, Copilot CLI, OpenCode and Google Antigravity, and every rule in the profile carries dated quotes from your own messages. The agent-driven flow writes separate files for work, design, writing and video, so a host can load only the one that fits the task. The repository behind this entry presents itself as Emulo. You can install it as a Claude Code or Codex plugin, bootstrap it with an npx skills command, or run the pip-installable CLI. The CLI also runs an MCP server. The agent-driven flow shows the cost and waits for approval before any model work. A free usage report counts how often you repeat yourself to a model and makes no model call, and an emulo verify command checks that every quote in a profile can be found in the mined sessions. Extraction, redaction, caches and the generated profile stay on your machine. Mining sends selected evidence to the model provider you choose, or stays fully local with a local model. The page is candid about limits: a published test found the mined profile did not separate from an invented one in one study, and in Claude Code the installed skill is not reliably opened on its own, so you type /you to load it.

Capabilities

Mines your own typed messages from Claude Code, Codex, Copilot CLI, OpenCode and Antigravity logs
Every rule backed by dated quotes, with a verify command that checks the quotes exist
Separate profile files for work, design, writing and video
Free usage report on repeated asks that makes no model call
Claude Code and Codex plugins, a pip-installable CLI and an MCP server

When to use it

You keep repeating the same preferences to your coding agent and want them written down from evidence
You want an inspectable profile with sourced quotes before anything reaches an agent
You want mining to stay on your own machine by pointing it at a local model