Browser automation CLI on raw CDP — no Puppeteer, Playwright or Selenium, 0 npm dependencies. 70+ commands, test assertions, click/fill by visible text without an LLM, and an MCP server for AI coding agents.

The most complete MCP server for API testing. 42 tools: HTTP requests (with token-optimized responses), assertions, request flows, OpenAPI import, mock data, load testing, collections, environments, native export/import, Postman import/export, cURL export

MCP server for Batfish: questions, assertions, topology, and failure testing as tools

Prompt evaluation workbench as an MCP server — run evals, compare providers, and test assertions directly from Claude Code or any MCP host.

A claims memory for agents: assert beliefs, let them decay, surface contradictions (MCP + CLI)

Monitor MCP servers, CI workflows, and HTTP endpoints with health history, TLS expiry, response assertions, and diagnostics

An MCP server that settles a claim about code instead of asserting it — runs a test with and without the fix it guards

MCP server that lets AI assistants analyze network configs with Batfish: 60 questions, assertions, failure diffs, and topology.

Sniff and decode BLE GATT traffic, fingerprint device profiles, and assert on insecure pairing/characteristics in CI against a capture.

A CRM your AI agent can actually run. MCP server for Claude, ChatGPT and Cursor: 45 tools, OAuth, and every value records who asserted it.

Agents make claims, Reelier writes receipts - record an agent workflow once, replay it deterministically at 0 tokens, and get an asserted receipt on every run.

Let AI agents test your iOS app in the Simulator — tap, type, swipe and assert, no testIDs needed. CLI + MCP server, zero dependencies.

MCP server: embedding pipeline ops + drift detection for production RAG. Assert query encoder matches index manifest; compute retriever stability between two index versions. Wraps embspec.

MCP server for testing: generate test cases, create mock data, mock API responses, analyze test coverage, generate assertions Install: npx @rog0x/mcp-testing-tools

Local memory, a codebase index, and working discipline for AI coding agents. Zero dependencies, no network, MCP included. Measured, not asserted.

MCP server: embedding pipeline ops + drift detection for production RAG. Assert query encoder matches index manifest; compute retriever stability between two index versions on a frozen probe set.

The verification layer for the agent economy: an agent-callable primitive that checks a number before it's asserted (evals, leaderboards, backtests), returns a signed reproducibility receipt, and meters itself (prepaid credits + x402 pay-per-call).

Local-first MCP server and CLI that audits test suites for false-green tests: tautological assertions, mock-contract drift, and mock hygiene. Read-only and deterministic, across TypeScript, PHP, Python and Rust.

Selenium MCP server for AI agents — 41 browser automation tools: navigation, clicks, typing, assertions, screenshots, multi-session, page snapshots with stable element refs, selector hints, and batched execution.

The most complete MCP server for API testing. 27 tools: requests, assertions, flows, OpenAPI, mock data, load testing, collections, environments, cURL export, response diffing. Zero config, zero dependencies.

Web navigation and verification for AI agents: a compact semantic page digest, a typed diff after each action, and assertions over that diff. MCP server and CLI on Playwright and SnapDOM.

Gherkio lets you describe HTTP-based integration tests as declarative YAML scenarios. Define requests, assertions, variable extractions, and orchestration — all in a simple, human-readable format that stays maintainable over time.

Local OCR and on-device screen reading for MCP clients (Claude, Cursor): read any window, locate elements to click, assert what's on screen. Apple Vision, no API keys, nothing uploaded.

One MCP server, 51 tools for AI agents on mobile + canvas UIs: iOS & Android automation, Maestro E2E, evidenced assertions, React Native/Metro debugging — plus a no-vision canvas/WebGL brain (Pixi/Konva/Fabric/Phaser/Three/Babylon) that drives game UIs like DOM elements, ~5x cheaper than screenshot loops.

Hand a whole browser task off in one call: a decision model drives the page server-side, so a flow costs one call, not a turn per click. Ref-based element tables, code-checked assertions, zero-model macro replay, over the Chrome DevTools Protocol.

Web-app QA, testing & analysis for AI agents — 74 Playwright tools: authenticated flows (2FA/SSO), 25-action E2E workflows, hard assertions, session reports, and accessibility / SEO / GEO / Core-Web-Vitals + Lighthouse audits across a page or a full crawled site. Battle-tested by an AI agent on real apps.

cdpilot is a browser automation CLI and local MCP server that talks to Chrome, Chromium, Brave or Vivaldi directly over the Chrome DevTools Protocol, with no Puppeteer, Playwright or Selenium and no browser download. Its 45 MCP tools cover navigation, clicks and form filling (including inside cross-origin iframes), accessibility-tree snapshots, screenshots, test assertions, a page's own WebMCP tools, and Web Bot Auth request signing. `cdpilot connect` attaches to a browser you started yourself, so a person can clear a login or CAPTCHA and the agent continues; cdpilot never closes that browser. Every feature has a runnable example with its real output: https://github.com/mehmetnadir/cdpilot/tree/main/examples

Web navigation, page observations and change verification for AI agents.

Judge hardware timing, serial boot logs and decoded traces against YAML contracts. pass/marginal/fail with evidence. MCP server, pytest plugin, GitHub Action.