emulo — Claude Code skill: profile and setup
emulo builds a personal working profile from local AI coding sessions, capturing coding preferences, design taste, writing style, and recurring mistakes for future tasks.
npx skills add ohad6k/emulo --skill clawhub -a claude-codeWhat emulo does
emulo examines local session logs from Claude Code, Codex, Copilot CLI, OpenCode, and Google Antigravity. It turns them into a you.md profile containing working rules, code and design preferences, writing habits, and failure patterns to avoid. Each rule comes with dated excerpts from actual sessions, so the profile is grounded in prior decisions rather than questionnaire answers.
The skill is intended to load this profile before an agent starts a task. To expose it through MCP, run uvx emulo mcp; the server provides load_emulo_profile for the work, design, write, and video domains. It returns an active profile or a recovery instruction when none is active. Running emulo --install you.md --target agents --repo . writes the profile into the project's AGENTS.md. That file is commonly committed and shared, so review its session excerpts and keep it out of version control when they should remain private.
Session mining happens locally, and the profile stays in the user's own file. The skill does not upload the mined data. The project previously used the name ditto-profile and was renamed Emulo in v0.5.0.
What it helps with
- extract coding rules from previous AI sessions
- carry personal design preferences into agent tasks
- record past failure patterns to avoid repeating them
- load a domain-specific profile for work, design, writing, or video
Example prompts for the agent
- Load my emulo profile before changing the code and follow the rules from my earlier sessions.
- Use my
designprofile when revising this interface. - Load my
writeprofile and match the writing style captured from my previous sessions.
How to install emulo
1Skills CLI (Vercel)
npx skills add ohad6k/emulo --skill clawhub -a claude-code2Manually: put the files in place
~/.claude/skills/clawhub/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/clawhub/
1Skills CLI (Vercel)
npx skills add ohad6k/emulo --skill clawhub -a codex2Manually: put the files in place
.agents/skills/clawhub/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add ohad6k/emulo --skill clawhub -a cursor2Manually: put the files in place
.cursor/skills/clawhub/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install ohad6k/emulo clawhub2Manually: put the files in place
.github/skills/clawhub/SKILL.md1Skills CLI (Vercel)
npx skills add ohad6k/emulo --skill clawhub -a gemini-cli2Platform CLI
gemini skills install https://github.com/ohad6k/emulo --path compat/clawhubManually: put the files in place
.devin/skills/clawhub/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add ohad6k/emulo --skill clawhub -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about emulo
How do I install emulo in Claude Code?
Run npx skills add ohad6k/emulo --skill clawhub -a claude-code. The skill file is located at ~/.claude/skills/clawhub/SKILL.md.
What does the emulo skill do?
It derives a working profile from local AI coding session logs. The profile records personal rules, preferences, and examples of past decisions that an agent can consult before working on a task.
How does emulo work with MCP?
Run uvx emulo mcp to start its MCP server. The load_emulo_profile tool returns the active profile for work, design, write, or video.
Source text
SKILL.md · MIT
---
name: emulo
description: Load the user's Emulo profile, rules mined from their own local AI coding session logs (Claude Code, Codex, Copilot CLI, OpenCode, Google Antigravity), each with dated quotes. Read it before working on their task.
emoji: 🐦⬛
homepage: https://github.com/ohad6k/emulo
metadata:
openclaw:
install:
- kind: uv
package: emulo
bins: [emulo]
---
# Emulo profile
Load the user's Emulo working profile: rules mined from their own coding
sessions, each with dated quotes. Read it before working on their task.
Emulo mines the user's own AI coding sessions (Claude Code, Codex, Copilot CLI,
OpenCode, Google Antigravity) into a `you.md` working profile: their laws, their
taste, their voice, and the failure modes they want you to protect them from.
## Use this before working on their task
If the profile is not installed yet, this writes it into the project's `AGENTS.md`
(it prints where it wrote and returns nothing to read). Do not run it yourself before
you name the exact folder to the user and get their yes. Tell them it writes the whole
profile, including quotes from their own sessions, into that AGENTS.md, which is usually
committed and shared, so suggest keeping it out of version control:
```
emulo --install you.md --target agents --repo .
```
Or call it through the MCP server:
```
uvx emulo mcp
```
The MCP server exposes one tool, `load_emulo_profile`, which returns the
profile for the domain you are about to work in (`work`, `design`, `write`, or
`video`). It serves only a profile activated by Emulo's agent mining flow; with
none active, it returns a recovery instruction.
## What you get
The profile is mined from the user's real sessions, not a questionnaire. It
carries:
- **Laws** they never want broken, each with receipts from real sessions
- **Taste** in design and code, including what they reject on sight
- **Voice** for anything written in their name
- **Failure modes** to protect them from
## Privacy
Mining runs locally against session logs already on the machine. The profile is
the user's own file. Nothing is uploaded by this skill.
Previously published as `ditto-profile`. The project was renamed to Emulo in
v0.5.0; see <https://github.com/ohad6k/emulo/releases/tag/v0.5.0>.
© ohad6k · MIT · ohad6k/emulo
