25 tagged agent testing, measured the same way as everything else here.
Browse within: agent-verification 14ai-testing-tool 14automated-testing 14playwright-mcp 14testing-tool 14AI Safety 5
Skill Claude CodeCodex
Reticle embeds a dev-only SDK in the user's running app and exposes it to you as reticle MCP tools. You look, act, observe, and assert against the real app. No screenshots, and no browser download for the verify loop: it drives the tab the user already has open.
Skill Claude CodeCodex
Install, instrument and verify this running web app from the inside (DOM, network, routing, console and framework state) instead of screenshots or guessing. Drives one real flow end to end and returns a verdict with the file:line to fix. Use when the user asks to set up or install Reticle, when a user-facing change…
Skill Claude CodeCodex
Verify that a web app change actually works by driving the running app from the inside (DOM, network, routing, console, framework state) instead of screenshots or guessing. Use after any user-facing change, when a fix is claimed but unproven, when a test passes but the UI is broken, or when you need a real verdict…
Skill Claude CodeCodex
Orchestrates end-to-end resilience testing for LLM agents with AgentBreak, including LLM infrastructure failures, prompt injection, agent skill supply-chain risk, guardrail verification, and MCP server/tool failures. Use when the user asks to "test my agent for resilience", "chaos test this agent", "find failure modes…
Skill Claude CodeCodex
Design, install, run, and report deterministic DeepSeek Harness tool-failure experiments. Use when a user wants to prove retry or fallback behavior, timeout or cooperative cancellation, policy-denial handling, blocked-result recovery, Code Mode nested-call resilience, or CI evidence for a DSH agent/plugin. Complete…
Skill Claude CodeCodex
Persistent multi-agent economy where autonomous AI agents compete for resources, trade on a marketplace, and benchmark decision-making against a standing population of always-on agents. Invite other agents for energy rewards. Auto-registers — no API key needed.
Skill Claude CodeCodex
Assist a QA engineer by navigating the app, drafting test cases from requirements, executing them, and recording results — with the human in the loop for judgment.
Skill Claude CodeCodex
Harden an LLM agent against a batch of FAIL findings from a single Vector session — group findings by root cause, propose minimal edits per group, add one regression test per group. Use after a full red-team session that produced multiple failures, or when the user says "fix all these findings".
Skill Claude CodeCodex
Generate an AgentContext JSON for Vector — interview the user about their LLM agent, produce a slug + 5-field JSON they paste into the cabinet's New Agent form. Use when creating a saved agent profile, or when the user asks "how do I describe my agent for Vector?".
Skill Claude CodeCodex
Integrate Vector red-team scanning into the user's codebase — build an AgentAdapter that calls their LLM agent, a RedTeamRunner that drives the Vector REST API, and a CI workflow that fails the build on Broken findings. Use when setting up Vector for the first time, wiring it into CI, or adding red-team gates to a…