Getting Started
Features
Everything EGC provides out of the box.
Core
Persistent memory
State is saved at the end of each session and restored at the start. Decisions, preferences, and next steps survive window closes and tool switches.
Global memory
A user-wide scope shared across every project: save transversal preferences and lessons once with scope global, and every session in every tool starts already knowing them. Project state always wins.
Token Crusher
Shell output is compressed up to 90% before it reaches the model: long git logs, test noise and install spam shrink while errors always survive. egc saved reports your accumulated savings at zero token cost.
Session Bus v2
Parallel sessions announce presence, split territory, take fail-fast cooperative path locks, and now exchange events through a per-session queue, with live presence surfaced automatically on every memory read.
Commit privacy
Populated memory never reaches a commit: pre-commit block, CI guard, and a git clean filter configured by egc init that stages a zeroed blob even when hooks are bypassed. Your context stays yours.
Bidirectional sync
update_state propagates context to all 11 tool config files automatically. Run egc watch to also sync changes you make directly in any tool file back to state and to all other tools in real time.
Multi-tool
Works with 20 AI coding tools. Install once and EGC registers itself in every tool you have installed.
Local-first
All state stays on your machine at ~/.egc/, encrypted at rest with AES-256-GCM. Nothing is sent externally.
Zero config
sh scripts/install.sh handles everything. Detects your tools, builds the servers, bootstraps the protocol.
Guardian layer
egc-guardian enforces safety at the harness level: every shell command and file write is validated before execution. UserPromptSubmit and PreToolUse hooks block dangerous patterns without AI behavioral calls.
Decision history
Every architectural decision is stored in SQLite with a timestamp. Query by date or tag.
Prompt library
Optional. Asked during installation. Written from real experience, not generated.
| Type | Count | What it is |
|---|---|---|
| Agents | 63 | Persona and behavior definitions |
| Skills | 230 | Domain-specific workflow runbooks |
| Commands | 77 | Command definitions and lifecycle hooks |
| Rules | 111 | Constraints and governance directives |
Persona and behavior definitions
Domain-specific workflow runbooks
Command definitions and lifecycle hooks
Constraints and governance directives
Organized per harness: works in .cursor/, .claude/, .gemini/, and more. Switch tools and the same workflows follow you.
Supported tools
| Tool | MCP | Bootstrap |
|---|---|---|
| Claude Code | ~/.claude/CLAUDE.md | |
| Antigravity CLI | ~/.gemini/GEMINI.md | |
| Gemini CLI | ~/.gemini/GEMINI.md | |
| Cursor | global cursor.rules | |
| Codex CLI | ~/.codex/config.toml | |
| OpenCode | ~/.opencode/instructions/EGC_MEMORY.md | |
| Windsurf | ~/.codeium/windsurf/ | |
| Amp | ~/.amp/ | |
| VS Code Copilot | ~/.github/ | |
| Zed | ~/.config/zed/AGENTS.md | |
| Continue.dev | ~/.continue/mcpServers/ | |
| Kiro | ~/.kiro/hooks/ | |
| Trae | context | ~/.trae/MEMORY.md |
| CodeBuddy | context | ~/.codebuddy/MEMORY.md |
| JetBrains Junie | context | .junie/guidelines.md |
| Goose | context | ~/.agents/skills/ |
| Amazon Q Developer | context | .amazonq/rules/ |
| OpenHands | context | ~/.agents/skills/ |
| Aider | context | .aider/skills/ + .aider.conf.yml |
| Warp | context | .warp/skills/ + AGENTS.md index |
~/.claude/CLAUDE.md
~/.gemini/GEMINI.md
~/.gemini/GEMINI.md
global cursor.rules
~/.codex/config.toml
~/.opencode/instructions/EGC_MEMORY.md
~/.codeium/windsurf/
~/.amp/
~/.github/
~/.config/zed/AGENTS.md
~/.continue/mcpServers/
~/.kiro/hooks/
~/.trae/MEMORY.md
~/.codebuddy/MEMORY.md
.junie/guidelines.md
~/.agents/skills/
.amazonq/rules/
~/.agents/skills/
.aider/skills/ + .aider.conf.yml
.warp/skills/ + AGENTS.md index
Supported LLM providers
EGC routes to any supported provider via ModelResolver. Fallback chains degrade automatically when a model is unavailable.
| Provider | Models |
|---|---|
| Google Gemini | Gemini 2.5 Pro / Flash / Flash-Lite, 2.0 Flash, 1.5 Pro / Flash |
| Anthropic Claude | Opus 4.5, Sonnet 4.7, Haiku 4.7 |
| OpenAI | GPT-4o, GPT-4o-mini |
| OpenRouter | DeepSeek R1, DeepSeek Chat V3, Qwen3 235B, Qwen3 32B, Llama 4 Maverick, Llama 4 Scout, Llama 3.3 70B, Gemini 2.5 via OR, GPT-4o via OR, and more |
| Ollama | Llama 3.2 (local, offline) |
Google Gemini
Gemini 2.5 Pro / Flash / Flash-Lite, 2.0 Flash, 1.5 Pro / Flash
Anthropic Claude
Opus 4.5, Sonnet 4.7, Haiku 4.7
OpenAI
GPT-4o, GPT-4o-mini
OpenRouter
DeepSeek R1, DeepSeek Chat V3, Qwen3 235B, Qwen3 32B, Llama 4 Maverick, Llama 4 Scout, Llama 3.3 70B, and more
Ollama
Llama 3.2 (local, offline)