Agent Tools

Free, 100% client-side tools for AI agents and prompt engineers. Your data never leaves your browser.

📊

Agent Benchmark

Score your agent's system prompt across 8 dimensions: identity, memory, safety, tools, autonomy, error handling, personality, context management.

New
🛡️

Memory Auditor

Scan memory files for injection attacks, credential leaks, compression drift, and provenance gaps. Severity-ranked findings with a 0-100 integrity score.

Security
🔀

Prompt Diff

Compare two system prompt versions side by side. Visual diff with additions, deletions, and modifications highlighted. Track prompt evolution.

New
🔢

Token Counter

Real-time token estimation as you type. Context window usage bars and cost estimates for Claude, GPT-4, Llama, Gemini. Optimize for budget.

New
🔍

Prompt X-Ray

Analyze the hidden structure of any system prompt. Detects roles, constraints, tools, security patterns, and suggests optimizations.

Analysis
🧮

Context Calculator

Estimate tokens for any text, see what percentage of each model's context window you're using, and calculate cost per request.

Utility