# FreeToolHub (remote · freetoolhub.org) 44 calculators for AI agents: US tax, finance, business + an MCP engineering & security suite. - Trust score: 62/100 (medium) - Registry status: active - Liveness: live - Owner verified: no - Last scored: 2026-09-20 ## Components - remote · `freetoolhub.org`: 62/100 (this document), [markdown](https://verifymcp.io/servers/cavan1977-freetoolhub/api-mcp.md), [page](https://verifymcp.io/servers/cavan1977-freetoolhub/api-mcp) ## Channel facts - Endpoint: `https://freetoolhub.org/api/mcp` - Transports: `streamable-http` - Auth: `none` - Version: `2.0.0` ## Trust breakdown How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog) Scored 2026-09-20. - **Endpoint Security**: 57/100 - The endpoint's TLS certificate is valid, in date, and uses a strong key. - Authorisation not fully verified: no authorisation is required to call this server, and 44 tool(s) never declared a destructiveHint. The MCP spec treats an absent hint as destructive by default, so we cannot call this surface safe. - HTTPS is enforced; there's no plaintext access path. - HSTS check failed: the Strict-Transport-Security header is absent. - DNSSEC check failed: this domain isn't protected by DNSSEC. - **Transport & Reachability**: 100/100 - Verified streamable-http transport via a live MCP handshake. - **Schema Quality & AI Usability**: 73/100 - AI-judged instruction clarity (good). - Context-footprint check failed: tool/resource definitions use about 4879 tokens (~110/item across 44 items; 44 tools + 0 resources), over budget; trim descriptions and params. - Usage-examples check failed: none of the tools include examples. - **Stability & Change Management**: 6/100 - Stability check failed: schema churn in the 5 days we've observed: 0 tool removals, 9 breaking changes, 0 auth/transport breaks, 0 additions. - **Tool Coverage**: 100/100 - 100% of tools have a non-trivial description (not blank, and not just the tool's name). - 100% of tool parameters carry a description. - **Tool Safety**: 100/100 - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured. - We read all 44 captured tool definition(s), and no name or description among them implies an irreversible operation. - An AI judge read all 44 captured unit(s) of tool text and found none that tries to manipulate the model reading it. - **Capabilities**: 20/100 - Spec-recency check failed: implements MCP spec 2024-11-05; the latest is 2026-07-28. ## Install ### How do I install the FreeToolHub MCP server? FreeToolHub is a hosted endpoint at https://freetoolhub.org/api/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation. ### Claude ```bash claude mcp add --transport http cavan1977-freetoolhub 'https://freetoolhub.org/api/mcp' ``` ### Cursor ```json { "mcpServers": { "cavan1977-freetoolhub": { "url": "https://freetoolhub.org/api/mcp" } } } ``` ### VS Code ```json { "servers": { "cavan1977-freetoolhub": { "type": "http", "url": "https://freetoolhub.org/api/mcp" } } } ``` ### Codex ```toml [mcp_servers.cavan1977-freetoolhub] url = "https://freetoolhub.org/api/mcp" ``` ### opencode ```json { "$schema": "https://opencode.ai/config.json", "mcp": { "cavan1977-freetoolhub": { "type": "remote", "url": "https://freetoolhub.org/api/mcp", "enabled": true } } } ``` ### OpenClaw ```bash openclaw mcp add cavan1977-freetoolhub --url 'https://freetoolhub.org/api/mcp' --transport streamable-http ``` ### Hermes ```yaml mcp_servers: cavan1977-freetoolhub: url: "https://freetoolhub.org/api/mcp" ``` ### Netclaw ```json { "McpServers": { "cavan1977-freetoolhub": { "Transport": "http", "Url": "https://freetoolhub.org/api/mcp" } } } ``` ### Vellum ```bash assistant mcp add cavan1977-freetoolhub -t streamable-http -u 'https://freetoolhub.org/api/mcp' ``` ### Other ```json { "mcpServers": { "cavan1977-freetoolhub": { "type": "http", "url": "https://freetoolhub.org/api/mcp" } } } ``` The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs. ## Changelog Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims. ### 2026-09-20 (score 62, +1) No change was recorded against any check on this day. Stability & Change Management went from 3 to 6. ### 2026-09-16 (score 61, 0) - [security regression] Stability: unverified → fail - [security regression] A breaking change shipped without a version bump: still 2.0.0 - [security] Tool “validate_mcp_server_card” rewrote its description, which is the text the model reads - [functional regression] “advise_tool_portfolio” changed the type of “tools”: string → string|array|object - [functional regression] “compress_tool_definitions” changed the type of “definitions”: string → string|array|object - [functional regression] “detect_polling_cost_traps” changed the type of “workflows”: string → string|array|object - [functional regression] “estimate_mcp_context_budget” changed the type of “config”: string → string|array|object - [functional regression] “generate_mcp_client_config” changed the type of “args”: string → string|array|object - [functional regression] “generate_mcp_client_config” changed the type of “env”: string → string|array|object - [functional regression] “generate_mcp_server_card” changed the type of “tools”: string → string|array|object - [functional regression] “plan_context_window_budget” changed the type of “servers”: string → string|array|object - [functional regression] “plan_mcp_token_budget” changed the type of “workloads”: string → string|array|object - [functional regression] “validate_mcp_server_card” changed the type of “card”: string → string|array|object - [cosmetic] “advise_tool_portfolio” reworded the description of “tools” - [cosmetic] “compress_tool_definitions” reworded the description of “definitions” - [cosmetic] “detect_polling_cost_traps” reworded the description of “workflows” - [cosmetic] “estimate_mcp_context_budget” reworded the description of “config” - [cosmetic] “generate_mcp_client_config” reworded the description of “args” - [cosmetic] “generate_mcp_client_config” reworded the description of “env” - [cosmetic] “generate_mcp_server_card” reworded the description of “tools” - [cosmetic] “plan_context_window_budget” reworded the description of “servers” - [cosmetic] “plan_mcp_token_budget” reworded the description of “workloads” - [cosmetic] “validate_mcp_server_card” reworded the description of “card” ### 2026-09-15 (score 61) First indexed and scored. ## MCP tools (44) ### `calculate_se_tax` (~77 tokens) Calculate US self-employment tax (15.3%), QBI deduction, quarterly payments. Uses latest IRS brackets from CDN data. Input parameters: - `filingStatus` (string): Tax filing status - `netEarnings` (number, required): Net SE income - `state` (string): US state abbreviation (e.g. CA, NY, TX) ### `calculate_freelancer_tax` (~91 tokens) Complete freelancer tax: SE tax, QBI deduction, federal + state income tax. Returns quarterly payment. Uses latest CDN data. Input parameters: - `businessExpenses` (number): Deductible expenses - `filingStatus` (string): Tax filing status - `grossIncome` (number, required): Gross freelance income - `state` (string): US state abbreviation (e.g. CA, NY, TX) ### `calculate_1099k_tax` (~113 tokens) Estimate tax on 1099-K income (reselling, gig work). SE tax, QBI, quarterly 1040-ES. Uses latest CDN data. Input parameters: - `costOfGoodsSold` (number): COGS - `filingStatus` (string): Tax filing status - `grossReceipts` (number, required): Total 1099-K receipts - `otherExpenses` (number): Other expenses - `state` (string): US state abbreviation (e.g. CA, NY, TX) ### `calculate_capital_gains_tax` (~120 tokens) US capital gains tax on stock/crypto. Short vs long-term, NIIT 3.8%, state tax. Uses latest CDN LTCG brackets. Input parameters: - `filingStatus` (string): Tax filing status - `holdingPeriodDays` (number, required): Days held (>365=long-term) - `ordinaryIncome` (number): Annual ordinary income - `purchasePrice` (number, required): Cost basis - `salePrice` (number, required): Sale price - `state` (string): US state abbreviation (e.g. CA, NY, TX) ### `compare_w2_vs_1099` (~102 tokens) Compare take-home pay: W-2 salary vs 1099 contractor rate. SE tax, FICA, standard deduction, state tax. Latest CDN data. Input parameters: - `contractorRate` (number, required): 1099 annual income - `filingStatus` (string): Tax filing status - `state` (string): US state abbreviation (e.g. CA, NY, TX) - `w2Salary` (number, required): W-2 annual salary ### `get_sales_tax_rate` (~63 tokens) Get state sales tax rate for any US state. Returns base state rate + estimated local rate. Latest CDN data. Input parameters: - `amount` (number): Purchase amount (optional) - `state` (string, required): US state abbreviation (e.g. CA, NY, TX) ### `calculate_payroll_tax` (~88 tokens) Calculate paycheck deductions: federal tax, Social Security, Medicare, state tax. All 50 states. Latest CDN data. Input parameters: - `filingStatus` (string): Tax filing status - `grossSalary` (number, required): Annual gross salary - `retirementContribution` (number): 401(k) contribution - `state` (string): US state abbreviation (e.g. CA, NY, TX) ### `calculate_estate_tax` (~62 tokens) Calculate federal estate tax. 2026 exemption $13.61M, 40% rate. Includes marital deduction. Input parameters: - `estateValue` (number, required): Total estate value - `spousalInheritance` (number): Amount left to spouse (marital deduction) ### `calculate_hsa_fsa_savings` (~112 tokens) Calculate tax savings from HSA and FSA contributions. Federal + FICA + state tax savings. Latest CDN limits. Input parameters: - `filingStatus` (string): Tax filing status - `fsaContribution` (number): FSA annual contribution - `hsaContribution` (number): HSA annual contribution - `marginalRate` (number): Your marginal tax rate % (e.g. 24) - `state` (string): US state abbreviation (e.g. CA, NY, TX) ### `calculate_quarterly_tax` (~90 tokens) Calculate quarterly estimated tax payments (Form 1040-ES). SE tax + income tax divided by 4. Latest CDN data. Input parameters: - `annualIncome` (number, required): Estimated annual net income - `businessExpenses` (number): Estimated expenses - `filingStatus` (string): Tax filing status - `state` (string): US state abbreviation (e.g. CA, NY, TX) ### `check_economic_nexus` (~62 tokens) Check which US states you have economic nexus in for sales tax collection. Based on Wayfair ruling. Latest CDN thresholds. Input parameters: - `statesWithSales` (number): Number of states you sell to - `totalSales` (number, required): Total annual US sales ### `calculate_tariff` (~86 tokens) Estimate import duties and landed cost by HS code category. Covers US MFN + Section 301 tariffs. Input parameters: - `category` (string): Product category - `freight` (number): Freight cost - `productValue` (number, required): Product value (FOB) - `tariffRate` (number): Tariff rate % (e.g. 7.5 for 7.5%) ### `calculate_compound_interest` (~94 tokens) Compound interest with monthly contributions. Returns final balance, total interest, inflation-adjusted value. Uses latest CDN inflation rate. Input parameters: - `annualRate` (number, required): Annual rate % (e.g. 7) - `inflationRate` (number): Inflation rate % (default: from CDN) - `monthlyContribution` (number): Monthly contribution - `principal` (number, required): Initial investment - `years` (number, required): Years ### `calculate_mortgage` (~120 tokens) Monthly mortgage payment (P&I) + property tax, insurance, PMI. Full PITI breakdown. Uses latest CDN mortgage rates for defaults. Input parameters: - `downPayment` (number, required): Down payment - `homePrice` (number, required): Home price - `insuranceRate` (number): Insurance % (default 0.35) - `interestRate` (number): Interest rate % (default: from CDN) - `loanTermYears` (number): Loan term years - `propertyTaxRate` (number): Property tax % (default 1.1) ### `compare_401k_vs_roth` (~119 tokens) Compare 401(k) pre-tax vs Roth IRA post-tax. 30-year growth, tax drag, recommendation. Uses latest CDN contribution limits. Input parameters: - `annualContribution` (number, required): Annual contribution - `annualReturn` (number): Annual return % (default 7) - `currentTaxRate` (number, required): Current tax rate % - `employerMatch` (number): Employer match $/yr - `retirementTaxRate` (number, required): Expected retirement tax rate % - `years` (number): Years to retirement (default 30) ### `calculate_car_loan` (~125 tokens) Monthly car payment, total interest, amortization. Includes sales tax, trade-in, down payment. Uses latest CDN sales tax rates. Input parameters: - `downPayment` (number): Down payment - `interestRate` (number, required): Interest rate % - `loanTermMonths` (number): Loan term months (default 60) - `price` (number, required): Car price - `salesTaxRate` (number): Sales tax % (default: from CDN by state) - `state` (string): US state for sales tax lookup - `tradeIn` (number): Trade-in value ### `calculate_rental_roi` (~104 tokens) Rental property ROI: cap rate, cash-on-cash return, NOI, annual cash flow. Includes vacancy and management. Input parameters: - `downPayment` (number, required): Down payment - `interestRate` (number): Mortgage rate % - `managementRate` (number): Property management % (default 8) - `monthlyRent` (number, required): Monthly rent - `propertyValue` (number, required): Property value - `vacancyRate` (number): Vacancy rate % (default 5) ### `calculate_student_loan` (~73 tokens) Student loan payoff: monthly payment, total interest, months saved with extra payments. Input parameters: - `extraPayment` (number): Extra monthly payment - `interestRate` (number, required): Interest rate % - `loanAmount` (number, required): Loan amount - `loanTermYears` (number): Loan term years (default 10) ### `calculate_roi` (~88 tokens) ROI, ROAS, net profit, CPA, CPC for marketing campaigns. Facebook, Google, TikTok ad analysis. Input parameters: - `adSpend` (number): Ad spend (for ROAS) - `clicks` (number): Ad clicks (for CPC) - `conversions` (number): Conversions (for CPA) - `cost` (number, required): Total cost - `revenue` (number, required): Total revenue ### `calculate_life_insurance` (~113 tokens) Calculate life insurance needs using DIME method: Debt, Income, Mortgage, Education. Input parameters: - `annualIncome` (number, required): Annual income (for income replacement) - `children` (number): Number of children - `collegeFund` (number): College fund per child - `debt` (number): Outstanding debt (excluding mortgage) - `mortgage` (number): Mortgage balance - `savings` (number): Existing savings/insurance - `yearsToReplace` (number): Years to replace income (default 10) ### `calculate_rental_affordability` (~54 tokens) How much rent can you afford? 30% rule, DTI ratio, comfortable rent range. Input parameters: - `monthlyDebts` (number): Monthly debt payments - `monthlyIncome` (number, required): Monthly income ### `calculate_home_equity` (~80 tokens) Calculate home equity, max borrowable amount (HELOC), and monthly payments. LTV limits included. Input parameters: - `homeValue` (number, required): Current home value - `interestRate` (number): HELOC rate % - `loanTermYears` (number): Loan term years (default 15) - `mortgageBalance` (number, required): Remaining mortgage balance ### `calculate_break_even` (~74 tokens) Break-even point: units to sell to cover fixed costs. Contribution margin, target profit analysis. Input parameters: - `fixedCosts` (number, required): Fixed costs - `pricePerUnit` (number, required): Price per unit - `targetProfit` (number): Target profit - `variableCostPerUnit` (number, required): Variable cost per unit ### `compare_llc_cost` (~63 tokens) Compare LLC formation and maintenance costs across US states. 5-year total cost. Uses latest CDN LLC state fees data. Input parameters: - `state` (string): State abbreviation (optional, omit for comparison) - `years` (number): Years to calculate (default 5) ### `calculate_dropship_profit` (~105 tokens) True dropshipping profit after product cost, platform fees, payment processing, ad spend. Uses latest CDN payment processor fees. Input parameters: - `adSpendPerOrder` (number, required): Ad spend per order - `paymentProcessor` (string): Payment processor ID - `platform` (string): Platform (shopify/etsy/woocommerce) - `productCost` (number, required): Product cost - `salePrice` (number, required): Selling price - `shippingCost` (number): Shipping cost ### `calculate_reseller_profit` (~82 tokens) Compare real profit across reselling platforms (eBay, Poshmark, Mercari, etc.) after fees. Uses latest CDN platform fee data. Input parameters: - `purchasePrice` (number, required): Purchase price - `salePrice` (number, required): Sale price - `shippingCost` (number): Shipping cost - `state` (string): US state for tax estimate ### `calculate_freelancer_pricing` (~81 tokens) Turn hourly rate into project quote with overhead, profit margin, milestone splits. Input parameters: - `hourlyRate` (number, required): Desired hourly rate - `overheadPct` (number): Overhead % (default 20) - `profitMarginPct` (number): Profit margin % (default 15) - `projectHours` (number, required): Estimated project hours ### `estimate_mcp_context_budget` (~156 tokens) Estimate how much of an LLM context window your MCP servers consume, and which tool definitions to cut first. Paste your MCP config. Returns per-server token cost, remaining budget, and ranked reduction advice. Input parameters: - `config` (string|array|object, required): Your MCP config as JSON. Either {"mcpServers":{...}} or a raw {name:{...}} map. Each server may declare "tools":[{name,description,inputSchema}]. Accepts a JSON object, or a JSON string containing on… - `contextWindow` (number): Model context window in tokens (default 200000) - `reservedForConversation` (number): Tokens you want to keep for the actual conversation (default 40000) ### `compress_tool_definitions` (~80 tokens) Shrink MCP tool descriptions and JSON schemas to cut context cost without changing how agents call them. Returns the compressed definitions plus exact token savings. Input parameters: - `definitions` (string|array|object, required): JSON array of tool definitions to compress. Each item: {name, description, inputSchema}. Accepts a JSON array, or a JSON string containing one. ### `score_mcp_server_trust` (~182 tokens) Score an MCP server A–F on trustworthiness and maintenance health, so you can choose between four servers that all claim to do the same thing. Supply what you know about the server. Input parameters: - `daysSinceLastCommit` (number): Days since the last commit - `hasInstallConfig` (boolean): Does it ship a working install config for Claude Desktop / Cursor / VS Code? - `hasToolDocs` (boolean): Does the README document the exposed tools and required credentials? - `inOfficialRegistry` (boolean): Is it listed in the official MCP registry? - `npmDeprecated` (boolean): Is the npm/PyPI package marked deprecated? - `repoExists` (boolean): Is the declared GitHub repo still reachable? - `serverName` (string): Server name (for the report) - `stars` (number): GitHub stars ### `audit_mcp_server_security` (~208 tokens) Audit an MCP server against the OWASP MCP Top 10. Reports authentication, path traversal, command injection, secret exposure and supply-chain risk with a concrete remediation list. Input parameters: - `acceptsUserInputInCommands` (boolean): Is untrusted input interpolated into commands or queries? - `executesCommands` (boolean): Does it execute shell commands or spawn processes? - `hasAuthentication` (boolean): Does the server enforce authentication on every request? - `hasHardcodedSecrets` (boolean): Are credentials hardcoded in the source or config? - `hasProvenanceAttestation` (boolean): Is the package published with build provenance / trusted publishing? - `performsFileOperations` (boolean): Does it read/write files based on model-supplied paths? - `restrictsNetworkEgress` (boolean): Are outbound network destinations restricted? - `serverName` (string): Server name (for the report) - `usesStdioTransport` (boolean): Does it run over local STDIO transport? ### `generate_mcp_server_card` (~161 tokens) Generate a SEP-1649 MCP Server Card (/.well-known/mcp/server-card.json) so crawlers and registries can discover your server without connecting to it. Input parameters: - `description` (string, required): What the server does - `endpointUrl` (string): Public Streamable HTTP endpoint URL - `name` (string, required): Server name - `requiresAuth` (boolean): Does the server require authentication? - `tools` (string|array|object): Optional JSON array of {name, description} for the capability summary. Accepts a JSON array, or a JSON string containing one. - `transport` (string): Transport: streamable-http or sse (default streamable-http) - `version` (string): Server version (semver) ### `validate_mcp_server_card` (~101 tokens) Validate an MCP Server Card against SEP-1649 and report every missing or malformed field before you publish it. Checks structure (serverInfo, transport, capabilities, authentication) and format (semver version, http(s) transport URL, capabilities.tools entries), returning a full error and warning list. Input parameters: - `card` (string|array|object, required): The server-card.json content to validate. Accepts a JSON object, or a JSON string containing one. ### `advise_tool_portfolio` (~103 tokens) Too many MCP tools makes agents pick the wrong one. Given your tool list and usage, get keep / merge / remove recommendations to get under the 10–15 tool guideline. Input parameters: - `targetToolCount` (number): Target number of tools to keep (default 12) - `tools` (string|array|object, required): JSON array of {name, category?, callsPerDay?, description?}. Accepts a JSON array, or a JSON string containing one. ### `plan_mcp_token_budget` (~155 tokens) Plan and attribute MCP token spend across agents, servers and workflows — something the protocol provides no native support for. Outputs per-dimension cost split and a quota plan. Input parameters: - `inputPricePerMillion` (number): Input price per 1M tokens in USD (default 3) - `monthlyBudget` (number): Optional monthly budget cap in USD for over/under comparison - `outputPricePerMillion` (number): Output price per 1M tokens in USD (default 15) - `workloads` (string|array|object, required): JSON array of {name, callsPerDay, inputTokensPerCall, outputTokensPerCall, owner?}. Accepts a JSON array, or a JSON string containing one. ### `generate_mcp_client_config` (~154 tokens) Generate ready-to-paste MCP server config for Claude Desktop, Cursor, VS Code and Windsurf from one input. Config completeness decides whether developers can install your server at all. Input parameters: - `args` (string|array|object): JSON array of command arguments. Accepts a JSON array, or a JSON string containing one. - `command` (string): Launch command for stdio servers (e.g. npx) - `env` (string|array|object): JSON object of environment variables. Accepts a JSON object, or a JSON string containing one. - `serverName` (string, required): Server name used as the config key - `url` (string): Endpoint URL for remote/streamable-http servers ### `find_tool` (~89 tokens) Semantic tool router. Given a natural-language intent, return only the most relevant tools from this server so the client can inject 3 schemas instead of 27. This is the "find_tool" meta-tool the MCP optimisation literature calls for. Input parameters: - `limit` (number): Maximum number of tools to return (default 3) - `query` (string, required): Natural-language description of what you want to accomplish ### `calculate_github_actions_cost` (~145 tokens) Estimate a GitHub Actions monthly bill using the January 2026 runner rates (Linux $0.006/min, Windows $0.010, macOS $0.062), accounting for plan allowances. Also reports what the same workload cost before the price cut. Input parameters: - `linuxMinutes` (number): Linux runner minutes per month - `macosMinutes` (number): macOS runner minutes per month - `plan` (string): GitHub plan: free, pro, team or enterprise (default free) - `publicRepo` (boolean): Are these workflows in public repositories? (unlimited free) - `windowsMinutes` (number): Windows runner minutes per month ### `compare_automation_platform_costs` (~96 tokens) Compare real monthly cost on Zapier (per task) versus Make (per operation) for the same workflow volume, including the counting-model difference that makes the two quotes non-obvious. Input parameters: - `runsPerMonth` (number, required): Workflow executions per month - `stepsPerRun` (number, required): Total steps (trigger + actions) per run - `workflowCount` (number): Number of distinct workflows (default 1) ### `detect_polling_cost_traps` (~115 tokens) Find the most common hidden cost in no-code automation: using polling where a webhook would do. Quantifies the wasted task/operation quota and the monthly dollars. Input parameters: - `platform` (string): zapier or make (default zapier) - `workflows` (string|array|object, required): JSON array of {name, triggerType: "polling"|"webhook", intervalMinutes, runsPerMonth, stepsPerRun, hasBranching?}. Accepts a JSON array, or a JSON string containing one. ### `estimate_selfhosted_tco` (~144 tokens) Compare paying hosted CI overage against running self-hosted runners, including the maintenance hours people forget to cost. Reports the true break-even volume. Input parameters: - `engineerHourlyCost` (number): Fully-loaded engineer cost per hour in USD (default 75) - `maintenanceHoursPerMonth` (number): Hours per month spent maintaining self-hosted runners (default 4) - `monthlyMinutes` (number, required): Current hosted runner minutes per month - `runnerHostingCostPerMonth` (number): Monthly infra cost for the self-hosted runner(s) in USD (default 20) - `runnerType` (string): linux, windows or macos (default linux) ### `compare_ai_coding_tool_costs` (~158 tokens) Compare monthly and annual cost of AI coding subscriptions (GitHub Copilot, Cursor, Claude Code/Pro/Max, OpenAI Codex) for a given team size and usage intensity, then compare the winner against pay-as-you-go API pricing for the same workload. Prices verified 2026-09. Input parameters: - `apiModel` (string): Model id for the API baseline (default claude-4-sonnet) - `includeApiBaseline` (boolean): Also compute the pay-as-you-go API cost for the same workload (default true) - `seats` (number): Number of developer seats (default 1) - `usageLevel` (string): Usage intensity: light, standard, heavy or extreme (default standard) ### `compare_llm_api_costs` (~179 tokens) Rank 18 LLM models by the real monthly cost of a given workload (input/output tokens per call × calls per month). Reports the cheapest and most expensive option, the cost spread, and which models can actually hold your prompt in their context window. Input parameters: - `callsPerMonth` (number, required): Number of API calls per month - `inputTokensPerCall` (number): Average input (prompt) tokens per call - `minContextWindow` (number): Only include models whose context window is at least this many tokens - `outputTokensPerCall` (number): Average output (completion) tokens per call - `provider` (string): Filter to one provider: openai, anthropic, google, deepseek or meta - `topN` (number): How many models to return, cheapest first (default 8) ### `plan_context_window_budget` (~212 tokens) Given a set of MCP servers and their tool counts, show for every model context window (128K, 200K, 1M, 10M) how much of the window the tool definitions consume, how much is left for conversation, and whether the server set fits at all. Answers "how many servers can I attach on model X?". Input parameters: - `avgTokensPerTool` (number): Default tokens per tool definition when a server omits it (default 550) - `conversationShare` (number): Fraction of the window you want left for human/agent conversation, 0-1 (default 0.25) - `reserveForOutput` (number): Tokens to reserve for the model response (default 8000) - `servers` (string|array|object, required): JSON array of {name, toolCount, avgTokensPerTool?} — e.g. [{"name":"github","toolCount":35}]. Accepts a JSON array, or a JSON string containing one. ## Diagnostics Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/cavan1977-freetoolhub/api-mcp#diagnostics ## Score history - 2026-09-20: 62 - 2026-09-19: 61 - 2026-09-18: 61 - 2026-09-17: 61 - 2026-09-16: 61 - 2026-09-15: 61 ## Common questions ### What is the FreeToolHub MCP server? FreeToolHub is an MCP server listed in the public MCP registry as io.github.cavan1977/freetoolhub. 44 calculators for AI agents: US tax, finance, business + an MCP engineering & security suite. This page covers its hosted endpoint (https://freetoolhub.org/api/mcp). ### Is the FreeToolHub MCP server safe to use? FreeToolHub scores 62 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm. ### What tools does the FreeToolHub MCP server expose? FreeToolHub exposes 44 tools: calculate_se_tax, calculate_freelancer_tax, calculate_1099k_tax, calculate_capital_gains_tax, compare_w2_vs_1099, and 39 more. Their descriptions and schemas cost roughly 4,879 tokens of context every time the server is loaded. ### Does the FreeToolHub MCP server require authentication? No. We connected to FreeToolHub without credentials and it answered, so anything it exposes is reachable by anyone who knows the address. ### Is the FreeToolHub MCP server still maintained? FreeToolHub is still listed as active in the MCP registry. We last reached this channel on 20 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced. ## Links - Remote endpoint: https://freetoolhub.org/api/mcp - Repository: https://github.com/cavan1977/freetoolhub - Website: https://freetoolhub.org/mcp/ - Changelog RSS feed: https://verifymcp.io/servers/cavan1977-freetoolhub/api-mcp.xml - Changelog JSON feed: https://verifymcp.io/servers/cavan1977-freetoolhub/api-mcp.json - HTML version of this page: https://verifymcp.io/servers/cavan1977-freetoolhub/api-mcp