Cross-agent context handoff protocol for AI coding agents

Give your AI a brain. Session memory, cross-session sync, and cross-agent context — one command per project.

Cross-session snapshot references and durable untrusted model context (ctx.sessionReferences)

The concrete agent loop plugin for the DeepSeek Harness

Per-call sandbox policy resolver and current model context: deployment fallbacks plus each session's mode and workspace root, shared by every enforcing capability family

Cross platform child_process#spawn and child_process#spawnSync

Generate and keep fresh cross-agent context files (AGENTS.md, SKILL.md) from your repo, deterministically. No LLM calls, no API key.

Linter for AI agent context files and MCP configs: CLAUDE.md, AGENTS.md, .cursorrules, .mcp.json - catches broken paths, wrong commands, leaked secrets

Zero-dependency browser-safe UUID and byte-encoding helpers

Agent interface, registry, initiator scope, and event vocabulary for the DeepSeek Harness

The standalone minimal SDK profile bundle: JSON-RPC, one DeepSeek adapter, persistent shell, editor, and JSONL sessions

Universal WHATWG Fetch API for Node, Browsers and React Native

Run scripts that set and use environment variables across platforms

Microsoft 365 Agents SDK for JavaScript

Capture the complete train of thought of agent work as first-class artifacts

Unified LLM API with automatic model discovery and provider configuration

A chat interface for AI coding agents — control Claude Code, Codex, Gemini CLI, OpenCode from browser and mobile

Opt-in durable per-step context with the current time and elapsed time

Type-only Branded<B> nominal-typing primitive for the DeepSeek Harness

User-approval seam (ctx.approval) for the DeepSeek Harness: one-shot permission decisions dispatched to composed answerers over the approval/request waterfall, fail-closed by default