atris

atris

atris is a self improving context layer for your agent. Works with coding agents like Claude Code and Cursor.

For Coding Agents

If you're a coding agent (Claude Code, Cursor, Windsurf, Copilot, etc.) reading this:

When the user asks to build/plan/fix something, run:

atris

Follow the instructions in the output. The full protocol is in atris.md.

Install

npm install -g atris

Quick Start

cd your-project
atris init    # Creates atris/ folder
atris         # Loads context, ready to build

Then describe what you want to build. The agent will: 0. (Optional) Explore with atris brainstorm if uncertain

• Show you a visualization
• Wait for approval
• Create atris/features/[name]/idea.md + build.md + validate.md
• Execute step by step
• Validate (fill in validate.md, harvest lessons if anything surprised you)

Commands: brainstorm (optional) → plan → do → review

Works with: Claude Code, Cursor, Windsurf, GitHub Copilot, any agent.

Experiments

Atris also supports Karpathy-style keep/revert loops inside atris/experiments/.

atris experiments init self-heal
atris experiments validate
atris experiments benchmark

Update

atris upgrade     # Install latest from npm
atris update      # Sync local files to new version

Skills

Atris includes agent-agnostic skills that work with Claude, Cursor, Codex, any LLM:

Skill	Purpose
atris	Workflow enforcement + plan/do/review
autopilot	PRD-driven autonomous execution
backend	Backend architecture anti-patterns
design	Frontend aesthetics policy
meta	Metacognition for agents
writing	Essay process with approval gates

Install to Codex: cp -r atris/skills/[name] ~/.codex/skills/

License: MIT | Repo: github.com/atrislabs/atris