MCP Server Marketplace: 950+ AI Tools
MCPize hosts 950+ Model Context Protocol servers for AI assistants, automation, databases, and search. One-click install for Claude, Cursor, VS Code, and any MCP client. One subscription gives you a single API key across every paid server.
Prevalid
Decision Intelligence for AI agents.
PulseNetwork Intel
Market and risk intelligence for AI agents from the PulseNetwork API fleet (76 origins, 950+ paid endpoints). Token-safety scans with CLEAR/CAUTION/AVOID verdicts for Solana memecoins and EVM tokens on 8 chains, official US macro indicators read from Chainlink oracles (GDP, PCE, SOFR), macro regime classification, ECB-reference FX conversion, country risk scoring, OFAC/EU/UN/UK sanctions exposure, and ACLED conflict scans. Deterministic sources, fast responses, no API keys to manage.
Human Emotion Intelligence
Human Emotion Intelligence provides developers with an open-source layer to integrate real-time emotion detection, intent analysis, and long-term emotional memory into LLM workflows. Use this server to build context-aware agents that adapt their response strategies based on user sentiment, enabling more empathetic and precise human-computer interactions.
Agent Reliability
Calculate success probability, bottlenecks and improvement tips for AI agent tool-chains. Pure math, zero API cost.
claude-cost-audit-mcp
Exact Claude API cost calculation with correct prompt-cache economics (1.25x/2x write, 0.1x read) and cache break-even analysis, plus a scanner that catches the single most common Claude-cost mistake — estimating tokens with OpenAI's tiktoken, which Anthropic's own docs say undercounts Claude by 15-20%.
regex-safety-audit-mcp
Parses a regex into a real AST to catch catastrophic-backtracking (ReDoS) risk — nested unbounded quantifiers, ambiguous alternation inside a repeated group, and backreferences — without ever executing the pattern itself. Generates proof-of-concept attack strings and JS-safe rewrites (JS has no atomic groups or possessive quantifiers, so the usual PCRE/.NET fixes don't transfer).
cron-schedule-audit-mcp
Validates cron expressions and computes correct next-fire times using the real IANA timezone database — catches the specific DST footgun that silently skips a scheduled job every spring or fires it twice every fall, plus the day-of-month/day-of-week OR-not-AND surprise almost everyone hits once.
mcp-schema-audit-mcp
Audits MCP tool definitions and marketplace listings for the specific patterns that make models call tools wrong, never call them, or mis-fill their parameters — vague descriptions, missing parameter docs, non-verb-first naming, oversized enums, deeply nested schemas — plus MCPize's own 500-character description limit, caught before you burn a deploy attempt finding out.
chrome-extension-submission-audit-mcp
Pre-submission Chrome Web Store policy audit for extension developers — manifest.json checks, remote-code/eval() scanning, and single-purpose/permission-justification checklists grounded in Google's own published policy categories (remote code, permission scope, single purpose, minimum functionality, privacy policy, misleading metadata). Catches the specific named reasons extensions get rejected for, before you submit and wait days for a form-letter rejection.
backtest-cost-sensitivity-mcp
Transaction-cost sensitivity analysis for backtested trading strategies — sweep commission/slippage assumptions to see exactly how fast an edge degrades, find the precise breakeven cost per trade, and run a structured look-ahead-bias audit. Answers the question a raw backtest report never shows on its own — how much of this edge is real friction-adjusted profit vs. an artifact of unrealistically cheap fills.
pinescript-audit-mcp
Static analysis for Pine Script v6 strategies — catches the specific known correctness traps that cause backtest-vs-live divergence (fill timing, ATR smoothing method, lookahead bias, VWAP session anchoring, exit-level drift, v6 syntax gotchas). Paste your strategy source, get a checked list back, not a compile — most of these bugs compile clean and only show up as silent numeric divergence.