MCPFlo: deterministic testing and debugging tool for MCP servers

Releases·Tool Release

Moka Chat gives Ollama models MCP tools via one npx command, with an inspector showing every tool call

A developer released Moka, a free, MIT-licensed local sandbox for testing prompts and MCP servers visually. It runs via 'npx @mokalabs/sandbox' with zero config, auto-detects running Ollama instances, and includes an inspector exposing raw JSON-RPC traffic, tokens, and latency. Developers can test tool calls and streaming, compare models side by side, and export working LangGraph Python scripts directly.

A developer released Moka, a free, MIT-licensed local sandbox for testing prompts and MCP servers visually. It runs via 'npx @mokalabs/sandbox' with zero config, auto-detects running Ollama instances, and includes an inspector exposing raw JSON-RPC traffic, tokens, and latency. Developers can test tool calls and streaming, compare models side by side, and export working LangGraph Python scripts directly.

Products·launched

Selfloop: visual tester that clicks through app flows

Selfloop is an open-source visual testing tool that automatically clicks through all flows of a web app to catch bugs. It uses LLMs to generate test scenarios and navigate the app, providing a faster alternative to manual testing or using Claude Code. It solves the problem of spending too much time testing vibe-coded apps.

Selfloop is an open-source visual testing tool that automatically clicks through all flows of a web app to catch bugs. It uses LLMs to generate test scenarios and navigate the app, providing a faster alternative to manual testing or using Claude Code. It solves the problem of spending too much time testing vibe-coded apps.

Researchers introduced AgentCheck, an open-source web workbench that allows developers to reproduce, intervene, and mitigate failures in tool-using LLM agents by injecting 12 types of faults into MCP server responses. It records real tool interactions, replays them with perturbations, and caches matching calls for efficient debugging before deployment.

Products·building

UI verification tool for non-deterministic generative UI in CI

uivet is an open-source tool that verifies generative UI outputs in CI pipelines. It samples multiple generations, renders them in headless Chromium, and scores data fidelity, accessibility, layout, console errors, and uses a multimodal judge with strict JSON schema enforcement. It solves the problem of testing non-deterministic UI outputs that change every render, which pixel-diff tools and text-only LLM evals cannot handle.

uivet is an open-source tool that verifies generative UI outputs in CI pipelines. It samples multiple generations, renders them in headless Chromium, and scores data fidelity, accessibility, layout, console errors, and uses a multimodal judge with strict JSON schema enforcement. It solves the problem of testing non-deterministic UI outputs that change every render, which pixel-diff tools and text-only LLM evals cannot handle.

Products·launched

uivet: CI tests for AI-generated UIs that vary at runtime

uivet is an open-source CI testing tool for interfaces that are dynamically generated by LLMs at runtime. It renders each scenario multiple times in a headless browser, checks data presence, runs accessibility audits, and uses a multimodal AI judge to score consistency. It solves the problem of testing unpredictable AI-drawn UIs for developers shipping LLM-generated interfaces.

uivet is an open-source CI testing tool for interfaces that are dynamically generated by LLMs at runtime. It renders each scenario multiple times in a headless browser, checks data presence, runs accessibility audits, and uses a multimodal AI judge to score consistency. It solves the problem of testing unpredictable AI-drawn UIs for developers shipping LLM-generated interfaces.