Viktor.comAn AI coworker that actually does the work
Launching today
Oynix
memory engine that actually remembers. built for AI agents.
1 follower
memory engine that actually remembers. built for AI agents.
1 follower
Your AI coding agent knows your code but not your company. Oynix gives it organizational memory, not just code search. It builds one live graph across repos, docs, Jira, ownership, incidents and past decisions, and serves it over MCP inside agnets like Claude Code, Codex and 16 others. Now agents answer cross-repo "why does this exist / what breaks / who owns this" from real company memory, on your own keys, instead of starting from zero every session.
Hi PH, today we're launching Oynix.
Index your codebase locally, and your coding agent becomes aware of all of it.
What you can expect:
1. One command indexes a repo or a folder on your machine. Nothing is uploaded.
2. Your agent queries it over MCP: what a symbol does, what calls it, who owns it, what breaks if you change it.
3. A per-module wiki, the repo hotspots, and a health score with dead code, dependency cycles and god objects.
4. Give it your intended architecture and it reports the imports that break it.
5. The agent writes down what it did, including the steps that broke, and reads it back next session. Start in Claude Code, continue in Codex.
Works with Claude Code, Cursor, Codex, Copilot, VS Code, Zed and a dozen more. macOS and Windows, free to install.
Give it a shot and tell us what your agent still couldn't answer, we're very keen on feedback!