Developer Tools1 month ago
Mirrors - Test AI Agents Against a Mirror of Production
Test AI agents against production mirrors using replayed traces to find bugs and regressions.
aisinghal
The full gallery
Tech stack
60 projects
Test AI agents against production mirrors using replayed traces to find bugs and regressions.
aisinghal
Buy and test specialized AI agents that automate tedious business and life tasks.
@coastallife831 · X
Everyone sells one AI chatbot. We sell 80 specialists you can test-drive instantly.
Design, validate, and compare AI agent deployments with governance controls.
@paulrodturner · X
Give AI agents a structured way to request human approval with safe retries and verification.
@GetAgentHail · X
AgentHail — a control layer for AI agents to request approval, execute work, and return verifiable results. Would you sign up or leave?
Debug and trace multi-agent AI pipelines with time-travel inspection.
suraj_chopade · Product Hunt
SwarmTrace Time-travel debugger for multi-agent AI pipelines
Chat with AI agents to scan web apps and APIs for security vulnerabilities.
@SableOffensive · X
Building Sable. An AaaS where specialized AI security agents perform penetration testing for web apps and APIs, helping teams find vulnerabilities before they reach production.
Autonomous agents that test web and mobile apps to discover flows, find bugs, and replay scenarios.
@AbdullahYusufY · X
Here is ours We are developing autonomous QA agents feel free to check it out.
Platform for building and deploying AI agents with isolated execution and cost metering.
@okiktech · X
Discover, purchase, and integrate AI agent capabilities via API.
b_radford · HN
Fund an agent once – prepaid USDC key for HTTP 402 tools (Solana)
Monitor, govern, and optimize your AI agents at scale.
@AiShivam · X
AgentStacKPro is an OS for AI agents designed to make them production-ready, and it holds incredible potential. You can also check out the project I currently developed
Evaluate AI products against published agent criteria with inspectable evidence and community votes.
@katyorby · X
i built — a local receipt for claude code runs. your check says whether the workspace passes now; the transcript supplies the activity counts. no transcript upload and no magical autonomy score.
Visual testing for you and your agents. An AI judge tells an intended change from a real regression, so you review decisions, not diffs.
@igorluchenkov · X
Hey Yuhao! Been running a lot of agents while building UI and built a tool for visual testing that integrates nicely with agents and gives your agent a way to make sure it didn't introduce any UI regressions: