Model access, evaluation and agent infrastructure.
OCD (Omni Coder Dashboard) — An AI memory engine with real semantic embeddings, 21 MCP tools, cross-tool routing, trace auditing, and proactive IDE interception. 100% local, zero API keys required.
Context Engineering Toolkit — generate GitHub Copilot customization files (agents, prompts, instructions, skills, hooks, MCP servers, agentic workflows) via a multi-agent AI pipeline with interactive CLI and 11 built-in templates.
Cross-agent skill manager — one global store, clean agent folders, activation via skill.config
MCP contracts for governed AI agent tools, OAuth scopes, and the Praxa agentic harness.
AI‑Agent‑Optimized CLI — structured output, predictable behaviour, zero prompts
MCP server for Alpine.js Template Outlet documentation - enables AI assistants to search and retrieve Template Outlet docs
Self-correcting 7-stage agent reasoning loop (SENSE→GKP→DECIDE+ACT→CHALLENGE-TEST→UPDATE→LEARN→REPORT). MCP server + CLI that scaffolds instruction files for GitHub Copilot, Cursor, and Claude Code.
Lumin observability + Agent Firewall for OpenClaw — captures prompts, responses, tool I/O, and synchronously enforces firewall policies on every tool call.
Sentinel AgentOS — 确定性 Guard 层 + 分层记忆 + 自动评估，让任何 Agent 变得可靠、可审计、可改进
Let Claude plan and validate coding tasks while OpenCode CLI, driven by a free model, does the implementation. CLI + MCP server.
Scan any MCP server for prompt injection, tool poisoning, leaked secrets, and destructive capabilities. One command, no key, no signup.
Type-safe, in-process testing toolkit for Model Context Protocol (MCP) servers. Test tools, resources, and prompts without subprocesses, with framework-agnostic assertions and a contract CLI.
Security scanner for MCP servers. Find vulnerabilities before your AI agent does.
MCP server for logging learning prompts with LLMs
MCP server for planning image-to-video shots and building production-ready AI video prompts
Local-stdio MCP server that scores how well your resume matches a job description. Runs on-device via Transformers.js — no API keys, nothing uploaded.
Capability-bounded execution for autonomous agents on Cowl. Let an agent trade from a shielded portfolio without ever holding a private key.
PM AID Full -- AI agent defense scanner with JS-native modules, refusal-bait scanner evasion, offline ED25519 licensing, Basic/Full module tiers, and zero-retention local scans.
Client-agnostic MCP server for Jira and Confluence — delivery intelligence, implementation prompts, and workflow automation for any AI coding agent
Skar turns a captured AI agent trace into a committed pytest regression test. MCP server + CLI. Use when a tool-using agent run fails and you want to lock the failure as an executable test.