209 tools found
Promptfoo workflow testing security weakness classification using Abliteration Large V2 model with configurable reasoning settings.
engrim is a local-first SQLite memory store that lets you switch between Claude Code, Cursor, Antigravity, and Windsurf without losing project decisions.
OpenAI Agents SDK workflow that routes Semgrep/Bandit scanner findings to specialist agents, validates them, then hands Codex a fix prompt.
Framework-free notebooks teaching production AI engineering from prompting to serving, with evals built in and verified on free APIs.
Open-source AI productivity OS from Cloudflare where every user gets a private, sandboxed app instance built by AI.
Multiplayer agent harness for startups, giving every employee and room its own scoped workspace in Slack and web.
Model-neutral MCP coding runtime: file edits, patches, command execution, and git for any MCP client.
Real-time detection, instance and semantic segmentation from one codebase - beats YOLO26/RF-DETR on Cityscapes F1 with fewer params.
Gateway injecting real API credentials into agent HTTP calls so agents only ever see placeholder keys, encrypted at rest.
Tamper-evident, hash-chained audit log for AI agent actions - verifiable by third parties without trusting the log's producer.
A six-phase security-audit skill finding exploitable, high-confidence vulnerabilities in authorized codebases, with strict anti-padding severity rules.
Reviews AI-generated WordPress code for security, core-API misuse, i18n, and performance issues before it ships, via a rule-based guard pass.