Consumer-agent surrogate testing for MCP servers, apps, browsers and APIs, with service-side outcome checks and sealed evidence
Agentic manual testing - Agentic Engineering Patterns
Agentic manual testing - Agentic Engineering Patterns
Property Testing with Agent Swarms
Using agents to build property-testing machinery, then turn the findings into small, reviewable fixes.
Writing Test Evals For Our MCP Server - Neon Blog
When we launched our MCP server, we knew it’d be important for it to have tests, just like any other piece of software. Since our MCP server has over 20 tools, it’s important for us to know that LLMs can pick the right tool for the job. So, this was the main aspect we wanted […]
Pluralistic: Google’s new remote attestation scheme is every bit as terrible as its old remote attestation scheme (12 Jun 2026) – Pluralistic: Daily links from Cory Doctorow
Long before "agentic AI," we had the idea that software would act as your agent on the internet. That's why the old-fashioned technical term for a browser is a "user agent." Your browser acts on your behalf to retrieve information and then show it to you, in the format you choose. It's your agent:
Web testing and quality for teams and AI agents | Cypress
Your AI writes the code. Cypress proves it works: end-to-end tests in a real browser, failures you can rewind, and coverage and accessibility on every run.
CrabTrap: Secure Agents in Production
CrabTrap is an LLM-as-a-judge HTTP proxy to secure agents in production. It intercepts and audits AI agent requests in real time. Try it on GitHub now.
evalstate/fast-agent
Code, Build and Evaluate agents - excellent Model and Skills/MCP/ACP Support
Google for Developers Blog - News about Web, Mobile, AI and Cloud
An open specification for finding and verifying tools, skills, and agents across the web.Agents are ...
Google for Developers Blog - News about Web, Mobile, AI and Cloud
An open specification for finding and verifying tools, skills, and agents across the web.Agents are ...
Agentic test processes, LLM benchmarks, and other notes on agentic coding from Galapagos Island
I've been using AI fairly heavily since last November and the whole thing is a funny experience. An agent will do something that, if a human did it, you'd immediately fire them. My reaction, of course, is to act as if this is great and spin up a thousand agents so they can do even more of that.
MCPJam/inspector
Testing and evaluation platform to chat, inspect, and debug MCP servers, MCP apps, and ChatGPT apps.
Test-Driven Approaches to Software Engineering with Large Language...
Tests increasingly participate in the decisions made by large language models and software engineering agents. They specify intended behavior, guide program construction and repair, select...
We won’t need to know
Two states of the agentic workflow: barebones architecture, and problems too complex to scrape.
Rethinking the Value of Agent-Generated Tests for LLM-Based Software Engineering Agents
Large Language Model (LLM) code agents increasingly resolve repository-level issues by iteratively editing code, invoking tools, and validating candidate patches. In these workflows, agents often write tests on the fly, a paradigm adopted by many high-ranking agents on the SWE-bench leaderboard. However, we observe that GPT-5.2, which writes almost no new tests, can even achieve performance comparable to top-ranking agents. This raises the critical question: whether such tests meaningfully improve issue resolution or merely mimic human testing practices while consuming a substantial interaction budget.