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.
Server has 4 tools with complete input schemas and descriptions. Naming follows verb_noun convention (run_crowd_test, list_personas, preview_mob, read_report). Descriptions are substantive (100-300 chars), explaining WHAT and WHEN. However, parameter descriptions lack constraint details (enums, ranges, formats). Output schema is not documented, LLMs cannot predict response structure. Error handling is minimal; no recovery guidance. Missing tool annotations (destructiveHint for run_crowd_test). No pagination support despite potential for large result sets.
List the built-in ringleader personas and what each one catches.
Preview the random mob members a given count/seed would generate.
Return the newest crowd-test markdown report under a directory. Defaults to ~/crowd-test-reports, where run_crowd_test writes.
Send a crowd of AI virtual users at a URL and return the damage summary. Only test sites the user owns or has permission to test. A run drives real browsers and typically takes several minutes; more personas and higher verify levels take longer.
Output schemas not documented. LLMs cannot predict response structure for any tool. build_summary() returns a string, but list_personas and preview_mob return unspecified objects.
Parameter constraints missing. 'verify' accepts 4 specific values but no enum declared. 'concurrency' and 'max_steps' lack min/max bounds. 'provider' should be enum, not free-form string.
No error recovery guidance. resolve_llm() raises RuntimeError with messages, but no tool-level error handling or recovery suggestions. LLM cannot self-correct on missing API keys or invalid provider.
Destructive operation lacks confirmation. run_crowd_test drives real browsers and writes reports, a high-impact action. No dry-run, confirmation step, or idempotent hint.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 56 | 2026-07-28+ | v2 |
Parameter descriptions lack actionable detail. 'seed' says 'Random seed for reproducibility' but does not explain valid range or default behavior. 'goal' says 'Extra mission' but does not clarify format or length limits.