free_tools
Tools I built to think with
Free, no sign-up. The same back-of-the-envelope models I use on real engagements, made interactive so you can run your own numbers.
reliability_scale
Reliability & scale
Set targets, size capacity, and put a number on the cost of going down.
SLO & Error Budget Calculator
Turn an availability target into the downtime it really permits, a request error budget, and the burn rate that says when to stop shipping and fix reliability.
Open toolDowntime Cost Calculator
Put a dollar figure on downtime: what an hour offline costs, your annual exposure, and what each additional 'nine' of availability is actually worth.
Open toolThroughput & Concurrency Calculator
Little's Law capacity planning: turn requests/second and latency into the concurrency you must support and the number of instances to serve it with headroom.
Open toolProduction-Readiness Scorecard
A 2-minute assessment across reliability, security, cost, scale and AI readiness, with a score and a prioritized list of what to fix first.
Open toolDockerfile & Cloud Run Auditor
Paste a Dockerfile for a graded production-readiness report tuned for Google Cloud Run: non-root, baked secrets, multi-stage, $PORT and signal handling, and image size, each with the exact line to change.
Open toolCloud Run Config Auditor
Paste a gcloud run deploy command or a Cloud Run service.yaml for a graded reliability and cost report: secrets via Secret Manager, max-instances bound, concurrency tuned to CPU and memory, CPU allocation mode, request timeout and the service account, each with the exact flag to change.
Open toolperformance
Performance
Budget latency and page weight, and turn speed into revenue.
Latency Budget Calculator
Set an end-to-end p95 target and carve it across the hops a request takes (network, app, database, cache) to see where the budget goes and what to fix first.
Open toolWeb Performance Budget Calculator
Set a total page-weight budget and spend it across HTML, CSS, JavaScript, images and fonts, add backend TTFB, and see the load time on a real connection. A page-weight budget for Core Web Vitals.
Open toolPage Speed → Revenue Calculator
Turn load time, traffic, conversion rate and order value into the revenue you lose to slowness, and what each 100ms is worth. The business case for a performance project.
Open toolRendering Strategy Picker
Answer four questions about freshness, personalization, SEO and scale, and get the right Next.js rendering strategy (SSG, ISR, SSR or CSR) with a config snippet and the trade-offs to watch.
Open toolCDN & Cache Savings Calculator
Turn your cache hit ratio, traffic and egress pricing into the dollars a CDN saves and the origin load it sheds, plus what five more points of hit ratio are worth.
Open toolai_agents
AI & agents
Estimate token cost and pressure-test whether an agent is production-ready.
Prompt Template Builder
Write a prompt with {{variables}}, fill in the blanks, and copy the finished prompt for ChatGPT, Claude or Gemini. Reuse one template across many situations.
Open toolAI Agent Cost Calculator
Estimate the monthly cost of an LLM agent across Claude, GPT-4o and Gemini, including the agentic context tax most calculators miss.
Open toolAI Agent Reliability Scorecard
Score whether your AI agent is production-ready across the loop-engineering disciplines: termination, escalation, prompt-injection defense, idempotency, context, cost, and observability.
Open toolAI Eval Suite Auditor
Paste an LLM eval suite (a promptfoo YAML or JSON test cases) for a graded report on whether it will actually catch a regression: deterministic runs, assertions that judge meaning not substrings, adversarial coverage, enough cases, and a pinned model.
Open toolLLM-as-Judge Rubric Linter
Paste your LLM-as-judge prompt for a graded report on the biases that make eval scores lie: missing criteria, position bias on pairwise comparisons, an undefined score scale, no reasoning before the verdict, length bias, and unparseable output.
Open toolAI Prompt Token & Cost Inspector
Paste a prompt for an exact token count from a real BPE tokenizer, the cost per call and per month across Claude, GPT-4o and Gemini, and how much of each model's context window it uses.
Open toolAI Agent Trace Analyzer
Paste an LLM agent trace (Anthropic Messages or OpenAI chat) to see turns, tool calls, the agentic context tax of re-sent tokens, and an estimated cost, with findings on loops, fat tool results and a bloated system prompt.
Open toolSystem Prompt Linter
Paste an LLM system or developer prompt for a graded report on the gaps that bite agents in production: missing stop conditions, no escalation path, prompt-injection surface, unconstrained output, and baked-in secrets, each with a concrete one-line fix. Runs in your browser.
Open toolLLM Tool Schema Generator
Paste a JSON Schema, an example JSON object, or a TypeScript interface and get a ready-to-paste tool/function definition for Anthropic, OpenAI and Gemini, with each provider's differences (strict required, additionalProperties, format, oneOf) flagged. Runs in your browser.
Open toolMCP & Agent Tool Auditor
Paste your LLM tool / function definitions (Anthropic, OpenAI, Gemini or an MCP tools/list) for a graded report on the failures that wreck tool-calling agents: tools the model can't tell apart, thin descriptions, undescribed or unconstrained parameters, and destructive tools with no guardrail, each with the exact fix. Runs in your browser.
Open toolMCP Server Security Auditor
Point it at a live MCP server (or paste its tools/list) for a STRIDE-style security report: prompt-injection text planted in tool descriptions, dangerous capability combinations that form an exfiltration path, over-broad inputs, and secrets exposed in the surface.
Open toolsecurity_seo
Security & SEO
Grade the headers and on-page signals search engines and attackers both read.
GitHub Actions Security Auditor
Paste a workflow YAML for a graded security report: unpinned actions, pull_request_target poisoned pipelines, script injection of untrusted input into run: shells, over-broad GITHUB_TOKEN permissions, and secrets handling, each with the exact line to change.
Open toolNext.js Config Auditor
Paste your next.config.js or .ts for a graded security and production report: missing security headers, secrets inlined into the client bundle via env, ignored type errors, an open image proxy, and public source maps, each with the exact line to change.
Open toolSecurity Headers Analyzer
Enter a URL for a graded report of its HTTP security headers (CSP, HSTS, X-Frame-Options, nosniff and more), with the exact header to paste for each gap.
Open toolSEO & Structured-Data Auditor
Enter a URL for a graded on-page report: title, meta description, canonical, indexability, Open Graph, Twitter cards, headings, viewport and JSON-LD structured data, with what to fix first.
Open toolAI Crawler & llms.txt Auditor
Enter a URL to grade how ready your site is for AI engines: whether GPTBot, ClaudeBot, PerplexityBot and friends are allowed, whether you ship an llms.txt, whether your content is readable without JavaScript, and the JSON-LD and sitemap that help you get cited.
Open toolGCP IAM Least-Privilege Analyzer
Paste a GCP IAM policy for a graded least-privilege report: public access, primitive roles (owner, editor, viewer), service-account impersonation, and over-broad admin grants, each with the tighter predefined role to use instead.
Open tooldev_utilities
Developer utilities
Everyday conversions and decoders that save a trip to the docs.
JSON to TypeScript & Zod
Paste any JSON and get clean TypeScript interfaces plus a matching Zod schema. Nested objects become named types, arrays of objects merge with optional fields, mixed types become unions. Runs in your browser.
Open toolcURL to Code Converter
Paste a cURL command and get the equivalent fetch, axios or Python requests code, with headers, JSON bodies and basic auth handled. Runs in your browser.
Open toolJWT Decoder & Inspector
Paste a JSON Web Token to decode its header and payload, read the claims, and see when it was issued and when it expires. Decodes in your browser; never verifies or stores it.
Open toolCron Expression Explainer
Paste a cron expression for a plain-English description and the next run times. Supports ranges, lists, steps, names and @macros.
Open toolOpenAPI Spec Linter
Paste an OpenAPI 3.x spec for a graded design-quality report: missing or unapplied auth, undocumented error responses, unpaginated collection endpoints, missing operationIds and response schemas, and unbounded arrays, each with the exact fix. Runs in your browser.
Open toolRegex Tester & Explainer
Test a regular expression against your own text, see every match and capture group highlighted live, and read the pattern back as a plain-English breakdown. Runs in your browser.
Open toolDiff Checker (Text & JSON)
Compare two blocks of text or two JSON documents and see the line-by-line additions and deletions. JSON mode normalizes formatting first so you get a semantic diff. Runs in your browser.
Open toolUUID & ULID Generator
Generate UUID v4, time-ordered UUID v7, or ULID identifiers in bulk, with copy buttons and a note on when each one is worth using for database index locality. Runs in your browser.
Open toolBase64 & URL Encoder/Decoder
Encode and decode Base64, URL-safe Base64, and URL components, with correct UTF-8 handling so unicode survives the round trip. Runs in your browser.
Open toolUnix Timestamp Converter
Convert a Unix timestamp to a readable date in UTC and local time, or turn a date back into seconds and milliseconds, with a live current epoch clock. Runs in your browser.
Open toolYAML to JSON Converter
Convert YAML to JSON and back, with live output and clear parse errors so you can untangle a Kubernetes manifest or CI config without a round trip to the docs. Runs in your browser.
Open toolGitHub Actions Workflow Generator
Describe your CI/CD workflow in plain English and get a ready-to-use YAML file, powered by AI. Or build one from presets for CI, deploy, release, preview, security scans, and dependency checks.
Open tool