AI-powered CLI coding agent with 20+ built-in tools, MCP support, and multi-model providers
This repository also configures its own agents. See what blade-code tells them →
Skill Claude Code
A permission and workspace-trust system for deciding whether tools, hooks, browser actions, project settings, and sensitive paths may be used. It supports allow, ask, and deny decisions and identifies trusted project sources.
Skill Claude CodeCodex
Shared runtime state and diagnostics for the command line, server, web app, agents, tools, and extensions. It tracks working directories, project roots, session IDs, logs, environment details, and registered projects.
Skill Claude CodeCodex
Shared request and response definitions for the server and web app, with runtime checks for incoming data and cleaned JSON schemas for tools. TypeBox is the library used to describe these data shapes.
Skill Claude CodeCodex
A workspace resource layer that gives each agent session its own fixed view of models, communication styles, hooks, and project instructions. It keeps the project that owns a resource separate from the working directory where files are edited.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: