Search

> A proof of concept to allow users with Overall/Read permission and Job/Configure (and optional Job/Build) to bypass the sandbox protection and execute arbitrary code on the Jenkins master or node.

Provider-agnostic AI guardrail testing framework. OWASP LLM Top 10 probe library.

Red-team your RAG pipeline for prompt injection and source-document leakage, in CI.

Authensor guardrail adapter for Vercel AI SDK

Probe a live URL for accidentally-served secret artifacts — .env, .env.production, .git/config, source maps (.js.map), .DS_Store, backup files — and CONFIRM each hit by fetching the bytes and flagging the real credentials and endpoints found inside. Zero

JavaScript API library for the ADAMANT Blockchain

Audit any Strapi CMS for public-role data exposure, CORS reflection, user enumeration and GraphQL introspection — and PROVE each leak live with an anonymous probe. No admin token needed for the keyless scan.

Async LLM jailbreak / prompt-injection fuzzer for OpenAI-compatible chat completion endpoints.

Mistral SDK v2 wrapper for BonkLM — chat/agents/fim/embeddings/classifiers guardrails with default-on multilingual + reformulation detection

Simple ports scanner for node.

Punish members who abuse admin functions

Vercel AI SDK callbacks for signing Signet tool-call receipts

A robust, production-ready CLI tool for detecting secrets and credentials in your codebase

Save a secret: a small window opens that Claude cannot see, you paste there, it lands in 1Password. Read it back: any item in your vault loads into Claude with one fingerprint check, names only, never the values. Keys you paste in the chat are deleted before Claude reads them. Mac, Windows, Linux. MIT.

BonkLM wrappers for Groq + Cerebras + Together inference providers (all OpenAI-compatible). Shared internal wrapOpenAICompatibleClient helper; 3 optional peer deps.

A solid TypeScript-based ESLint plugin infrastructure inspired by typescript-eslint

Canonicalize before you guard: a pure-stdlib Unicode normalizer for LLM prompt-injection defense (invisible-character stripping, homoglyph mapping, locale-independent case folding).

Turkish and multilingual prompt-injection / jailbreak probes for NVIDIA garak (LLM vulnerability scanner).

Guarded vector-database client wrapper for BonkLM — validation, secret and injection guardrails around a self-hosted vector search engine

ESLint rules that AI assistants can actually fix - error messages optimized for both human developers and Large Language Models