Workflow Tools
Free browser-based tools for building, testing, and optimizing AI workflows. Design prompt chains, manage token budgets, and track costs.
Visual Workflow Designer
Drag-and-drop workflow builder with JSON export. Chain AI prompts into visual flowcharts with input, condition, transform, and output blocks.
Context Window Optimizer
Token budget planner and chunking strategy tool. Optimize how you allocate tokens across system prompt, context, and response in your workflow.
System Prompt A/B Tester
Compare two system prompts side by side with response quality scoring. Test different prompt strategies to find what works best for your use case.
Claude API Cost Calculator
Estimate Claude API spend from input and output token counts. Four models, per-token rates, prompt-caching pricing, and a batch-vs-streaming view, computed in your browser.
Prompt Caching Savings Estimator
Estimate what prompt caching saves per request on Claude Opus, Sonnet and Haiku. Shows cache-write and cache-read math per call with the formula and assumptions shown, computed in your browser.
Batch API Cost Savings Calculator
Compute the exact 50% Batch API discount on Claude Sonnet, Opus and Haiku from your request mix. Shows sync vs batch monthly cost, dollars saved, and what still must run synchronously, computed in your browser.
Haiku vs Sonnet Router Savings
Model the monthly savings from routing easy Claude traffic to Haiku, including the retry share that has to escalate back to Sonnet and the router overhead tokens. Break-even retry rate shown, computed in your browser.
MCP Tool Result Token Budget
Plan how many tokens each MCP tool result may occupy in a Claude context window. Deducts system prompt, conversation, and answer headroom, then splits the remainder across tool calls, computed in your browser.
background:var(--bg-card);border:1px solid var(--border);border-radius:var(--radius);padding:1.5rem;text-decoration:none;transition:border-color 0.2s;">Opus vs Sonnet Turn Cost
Turn-by-turn cost comparison for Claude Opus and Sonnet agent runs. Compares first-turn, last-turn, and total run cost from Anthropic list pricing, and shows how many extra turns Opus can afford.
Token Cost Tracker
Claude API usage dashboard with budget alerts. Track token consumption, estimate monthly costs, and set spending thresholds for your workflows.
Cache Invalidation Cost Multiplier
A timestamp in your system prompt makes every request rewrite the cache. Calculate the cache invalidation cost multiplier - up to 12.36x on Anthropic.
Agent Team Cost per Deliverable
A 5-agent team with human review costs more than one agent plus an editor. Model the true per-deliverable cost of your agent team.
Parallel Session Cost Blower
Six parallel Claude Code sessions burn 6x the tokens and drain your rate cap 6x faster. Calculate the real monthly bill before you spawn the fleet.
Prompt Prefix Refactor Savings
Move dynamic content after the static prefix and cut input cost by up to 84%. Calculate the savings from prompt prefix reordering.