{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_5ascrvsy3nep","handle":"crowd-test-mcp","url":"https://wellknown.network/agents/crowd-test-mcp","links":{"self":"https://wellknown.network/agents/crowd-test-mcp/record.json","html":"https://wellknown.network/agents/crowd-test-mcp","markdown":"https://wellknown.network/agents/crowd-test-mcp/record.md","api":"https://wellknown.network/api/v1/agents/crowd-test-mcp","status":"https://wellknown.network/api/v1/agents/crowd-test-mcp/status","claim":"https://wellknown.network/agents/crowd-test-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/crowd-test-mcp/claim.json","badge":"https://wellknown.network/agents/crowd-test-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/crowd-test-mcp/history","tools":"https://wellknown.network/api/v1/agents/crowd-test-mcp/tools"},"ard":{"identifier":"urn:air::server:crowd-test-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"crowd-test-mcp","summary":"MCP server for crowd-test: unleash a mob of AI virtual users on your website from Claude Desktop, Claude Code, Cursor, or any MCP client.","description":"# crowd-test-mcp 🔥\n\n**The mob, on tap — an MCP server for [crowd-test](https://github.com/anhhuyn411-alt/crowd-test).**\n\nAsk Claude Desktop, Claude Code, Cursor, or any MCP client to unleash a crowd\nof AI virtual users — impatient shoppers, confused seniors, keyboard-only\nusers, chaos monkeys — on your website. They browse it in real Chromium,\nfile findings, and hand back a damage report with a **survival grade** (S–F).\n\n> *\"Send the mob at https://staging.myapp.com and tell me what to fix first.\"*\n\nThat's the whole workflow now.\n\n## Install\n\n```bash\npip install crowd-test-mcp\n```\n\nAn LLM key is required in the server's environment: `ANTHROPIC_API_KEY` or\n`OPENAI_API_KEY`.\n\n### Claude Code\n\n```bash\nclaude mcp add crowd-test -e ANTHROPIC_API_KEY=sk-... -- crowd-test-mcp\n```\n\n### Claude Desktop / Cursor / anything MCP\n\n```json\n{\n  \"mcpServers\": {\n    \"crowd-test\": {\n      \"command\": \"crowd-test-mcp\",\n      \"env\": { \"ANTHROPIC_API_KEY\": \"sk-...\" }\n    }\n  }\n}\n```\n\n## Tools\n\n| Tool | What it does |\n|---|---|\n| `run_crowd_test` | Send the crowd at a URL. Pick personas, add a random `mob`, set a `goal`, choose the verification depth (`none` / `detective` / `cross` / `tribunal`). Returns a compact damage summary; full markdown/HTML reports land in `~/crowd-test-reports/`. |\n| `list_personas` | The ten built-in ringleaders and what each one catches. |\n| `preview_mob` | Preview the random mob a given `count`/`seed` would generate. |\n| `read_report` | Fetch the newest full markdown report from disk. |\n\n## The verification tribunal\n\nFindings can be cross-examined by up to three independent harnesses before\nthey count against the grade — a skeptical detective agent, a raw Playwright\nprobe, and a [Microsoft Webwright](https://github.com/microsoft/Webwright)\nagent. Automation artifacts get disputed instead of panicking you. Details in\nthe [crowd-test README](https://github.com/anhhuyn411-alt/crowd-test#the-tribunal-%EF%B8%8F).\n\nDeeper layers need one-time extras:\n\n```bash\npip i…","publisher":null,"homepage":"https://github.com/anhhuyn411-alt/crowd-test-mcp","repository":"https://github.com/anhhuyn411-alt/crowd-test-mcp","version":"0.1.0","license":null,"protocols":["mcp"],"tags":["ai-agents","browser-automation","mcp","qa","testing","ux"],"pricing":null,"endpoints":[{"url":"pypi:crowd-test-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","repoUrl":"pypi","summary":"pypi","version":"pypi","description":"pypi","homepageUrl":"pypi"}},"derived":{"capabilities":[{"slug":"code.testing","name":"Testing","confidence":1,"provenance":"declared"},{"slug":"infra.browser-automation","name":"Browser Automation","confidence":1,"provenance":"declared"}],"categories":["code","infra"],"language":"en"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"crowd-test-mcp","registry":"pypi","observedAt":"2026-10-27T22:27:55.948Z","publishedAt":"2026-07-23T04:34:19.699329Z","latestVersion":"0.1.0"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"crowd-test-mcp","url":"https://pypi.org/project/crowd-test-mcp/","firstSeenAt":"2026-09-09T13:20:50.563Z","fetchedAt":"2026-10-27T22:27:10.907Z","normalizedAt":"2026-10-27T22:27:10.907Z"}]},"firstSeenAt":"2026-09-09T13:20:50.563Z","updatedAt":"2026-10-27T22:27:55.948Z"}