Releases·Tool Release
Jango launches a Mac app for testing multi-user flows with AI agents
Jango launched a Mac app that gives each AI agent participant its own browser, account, goal, and memory, letting developers test workflows that need more than one user. Runs can be watched, directed, or joined by a human, and produce action logs, checks, screenshots, and reusable app knowledge. It targets a gap ordinary end-to-end tests miss: state and permissions that only emerge when several accounts interact in the same app.
3 engagement1 sourceSat, Sep 26, 2026, 07:50 PM
AI Overview
- Jango gives each AI agent its own browser, account, goal, and memory to test multi-user workflows like social feeds, group chat, and marketplaces.
- Humans can watch, direct, or join runs; each run leaves action logs, checks, screenshots, and reusable app knowledge.
- The app connects via CLI or MCP and runs locally on Mac only, a limitation for other platforms and strict test-data controls.
Related
Products·launched
Agent memory and loop tracking tool
A tool that helps developers monitor and debug AI agents by tracking their actions, preventing loops, and managing memory. It allows agents to share memory across different functions (e.g., customer support agent talking to pricing agent). Built for daily agent users who need visibility into agent behavior.
A tool that helps developers monitor and debug AI agents by tracking their actions, preventing loops, and managing memory. It allows agents to share memory across different functions (e.g., customer support agent talking to pricing agent). Built for daily agent users who need visibility into agent behavior.
0 engagement1 sourceredditMon, Jul 13, 2026, 10:46 AM
Products·launched
AgentSnap: Snapshot testing for AI agents
AgentSnap is a tool for snapshot testing of AI agents, allowing developers to capture and compare agent outputs across runs. It uses LLMs to generate deterministic snapshots of agent behavior, helping debug regressions and ensure consistency. The tool targets developers building complex agentic systems who need reliable testing.
AgentSnap is a tool for snapshot testing of AI agents, allowing developers to capture and compare agent outputs across runs. It uses LLMs to generate deterministic snapshots of agent behavior, helping debug regressions and ensure consistency. The tool targets developers building complex agentic systems who need reliable testing.
3 engagement1 sourcehackernewsWed, Jul 29, 2026, 04:18 PM
Products·launched
API for computer-use agents that operate legacy desktop and web apps
Coasty provides an API and consumer app for computer-use agents that complete workflows inside legacy desktop software and web applications without usable APIs. Developers send natural-language tasks, select a machine or browser environment, and the agent operates the interface via screenshots, mouse, and keyboard, returning a structured run record.
Coasty provides an API and consumer app for computer-use agents that complete workflows inside legacy desktop software and web applications without usable APIs. Developers send natural-language tasks, select a machine or browser environment, and the agent operates the interface via screenshots, mouse, and keyboard, returning a structured run record.
20 engagement1 sourcehackernewsWed, Jul 15, 2026, 03:51 PM
Products·launched
Open-source MCP system for AI agents to write and manage Playwright browser automations
An open-source platform that connects to AI agents via MCP so they can write and manage Playwright browser automations for testing or workflow automation. It adds scheduling, notifications, IAM, and hooks around those automations. A hybrid Chrome extension lets agents read the page to generate automations directly instead of relying on written prompts.
An open-source platform that connects to AI agents via MCP so they can write and manage Playwright browser automations for testing or workflow automation. It adds scheduling, notifications, IAM, and hooks around those automations. A hybrid Chrome extension lets agents read the page to generate automations directly instead of relying on written prompts.
1 engagement1 sourcehackernewsSat, Sep 12, 2026, 10:00 PM
Products·launched
Dynobox: Test runner for AI agent skills and workflows
Dynobox is a test runner designed to validate AI agent skills and workflows. It helps developers ensure their AI agents perform correctly by running automated tests. The tool addresses the need for reliable testing in AI agent development.
Dynobox is a test runner designed to validate AI agent skills and workflows. It helps developers ensure their AI agents perform correctly by running automated tests. The tool addresses the need for reliable testing in AI agent development.
2 engagement1 sourcehackernewsWed, Aug 12, 2026, 04:14 PM