Atris functions as an operating system for AI-assisted development by turning any codebase into a persistent, agent-ready workspace. It provides cross-session project memory for tools like Claude Code, Codex, and Cursor using plain Markdown files and SQLite tasks within the repository. Following a strict plan-do-review workflow, Atris verifies that agent tasks conclude with passing tests and verifiable proof rather than unverified completions. With support for autonomous multi-agent dispatch, answer engine optimization checks, and automated pull request landing, it coordinates agents and humans within a shared operational context.
主要功能
01Self-landing mechanisms that automate branch pushing, PR generation, and GitHub auto-merge
02Durable cross-session project memory via plain repository files and structured wikis
03Integrated Answer Engine Optimization (AEO) scanner for LLM visibility and llms.txt drafting
0471 GitHub stars
05Autonomous multi-agent dispatch and execution with separate reviewer validation
06Verification-driven workflow using a plan, do, and review loop backed by test runs
使用案例
01Automating routine code fixes and refactoring tasks with mandatory proof-of-work tests
02Equipping coding agents with persistent codebase memory across multiple sessions and tools
03Auditing and optimizing website readability for AI search engines like ChatGPT, Claude, and Perplexity