Search

OpenCode plugin that detects prompt injection in tool call outputs using an LLM judge

A modern network scanner wrapping Nmap, designed for intuitive network reconnaissance.

JS SDK for Cigalpay API Integration

An ML-KEM/CRYSTALS-KYBER implementation written in TypeScript for various JavaScript runtimes

OAuth 2.0 Client Plugin for Catberry Framework

Deterministic password, SSH key, and wallet derivation from a master secret

True Attribute-Based Access Control (ABAC) engine for Node.js - Authorization decisions based purely on attributes

A claude code proxy to sniff the web requests it makes in a session and allow users to block suspicious calls.

<div align="center">

A Bcrypt hash generator & checker

Run npm audit and fail on high/critical vulnerabilities unless explicitly accepted

ESLint parser that allows for linting of experimental syntax transformed by Babel (with updated dependencies)

AI-Proof File Guard - Protect sensitive files from accidental AI modifications

Sandboxed Claude Code, Codex and Pi: a bubblewrap jail with an egress allowlist, via uvx

AI breach defense package — protect any website from AI-driven attacks

Offline, in-process input/output safety guard for LLM apps (PII vault, prompt-injection, secrets, malicious URLs, package hallucination). Rust core, Node (napi) + edge/browser (WASM). Defense-in-depth, not a guarantee.

Zero-latency compliance auditing for every LLM call in your application

Express.js middleware for tracking and blocking abusive behaviour.

TinyForest is dedicated to providing eco-friendly, small-scale forest solutions for urban environments, ensuring lush greenery even in the heart of the city. Our product line combines innovative planting techniques with sustainable practices, allowing users to create dense, biodiverse forest patches in limited spaces. TinyForest not only transforms the aesthetic appeal of urban areas but also contributes to cleaner air, better microclimates, and a significant boost to local biodiversity. With ou

Filter OpenCode skills on a per-agent basis using configurable allowlists and blocklists