Fullstack developer at NetGenium in Prague. Day to day I work on RAG pipelines, MCP servers and process automation for enterprise clients; in my free time I build open-source tooling for AI agents.
🌐 ondrejulehla.dev · 💼 LinkedIn
- readme-ci – run the code blocks in your README and fail CI when your quickstart breaks
- agent-audit – security auditor for AI agent configurations: MCP servers, Claude Code settings, skills and hooks
- agent-lens – drop an agent session transcript and get a visual debrief: tool-call timeline, tokens, cost, failures
- mcp-sandbox – run any MCP server inside an isolated E2B cloud sandbox with one command, no code changes
Recent PRs in the AI agent ecosystem:
- e2b-dev/E2B – fix(cli): resolve template alias instead of writing template_id as name in migrate
- e2b-dev/code-interpreter – fix(js): linear buffering and stateful UTF-8 decoding in readLines
- e2b-dev/surf – fix: decouple sandbox resolution from browser breakpoints
- e2b-dev/e2b-cookbook – add a Claude Agent SDK in E2B Sandbox example (TypeScript)
- langchain-ai/langchainjs – fix(google-genai): serialize standard v1 reasoning content blocks
- anthropics/claude-agent-sdk-python – docs: document env var inheritance into the CLI subprocess