30-Day Free Trial
Claude Opus 4.7●DeepSeek V4-Pro●GPT 5.5●Gemini 3 Ultra●Llama 4.5 Behemoth●Grok 4.5 Aurora●OpenAI o5●R1 1776●Claude Opus 4.7●DeepSeek V4-Pro●GPT 5.5●Gemini 3 Ultra●Llama 4.5 Behemoth●Grok 4.5 Aurora●OpenAI o5●R1 1776●
Claim Now →

Prompt Library

Prompts tuned for our stack

Free 30 days · No credit card

Production-grade prompts engineered against the models we ship. Each one shows its winrate and the sample size behind it — what you see is the accepted-diff rate across thousands of real regression runs. Click a prompt to copy the full body.

Prompt Engineering Guide

OWASP Top 10 security audit with remediation plan

Full security scan with ranked findings and fix suggestions

Claude Opus 4.794% winrate· over 18k runs

Secrets scanner — detect API keys, tokens, passwords

Scan entire repo for hardcoded credentials and PII

Claude Sonnet 4.792% winrate· over 12k runs

Dependency CVE audit with safe upgrade paths

Check npm/pip/go deps against CVE databases

Claude Sonnet 4.789% winrate· over 9.1k runs

SQL injection test case generator

Generate comprehensive SQLi test cases from schema

GPT-5 Pro88% winrate· over 8.7k runs

XSS payload generator for WAF bypass testing

Fuzz test XSS filters with encoding variations

Claude Sonnet 4.786% winrate· over 7.3k runs

Container security scan via Trivy

Analyze Dockerfile for CVEs and misconfigs

GPT-591% winrate· over 6.2k runs

JWT token security audit

Check JWT implementation for common vulnerabilities

Claude Sonnet 4.790% winrate· over 5.8k runs

CORS policy evaluator

Audit CORS headers and suggest secure configurations

GPT-584% winrate· over 4.9k runs
57+
Prompts shipped
87%
Avg winrate
511k
Regression runs
3h
Re-test interval

Chainable workflows

Prompts work together

prompt chain
# security-audit → fix-issues → generate-tests
$ /prompt run security-audit
✓ Found 3 high, 7 medium severity issues
$ /prompt run fix-issues --severity=high
✓ 3 files auto-fixed
$ /prompt run generate-tests --coverage
✓ 12 new test cases added
Want to write your own?

Read our Prompt Engineering Guide.

Six authoring principles, a canonical template, the regression harness that keeps every prompt honest, and the workflow that gets your prompt into this library. Written by the team, for the team and the people who join next.

6 principlesCanonical templateAuthoring workflowRegression harness
Open the guide