EdgeCrab 🦀

"Your SuperAgent — built in Rust."

EdgeCrab is a SuperAgent — a personal assistant and coding agent forged in Rust. It carries the soul of Nous Hermes Agent (autonomous reasoning, persistent memory, user-first alignment) and the always-on presence of OpenClaw (15 messaging gateways, smart-home integration), compressed into a 15 MB static binary that starts in < 50 ms with zero runtime dependencies.

Architecture

hermes-agent soul  +  OpenClaw vision  =  EdgeCrab
   (reasoning)          (presence)        (Rust)

| Metric | EdgeCrab 🦀 | hermes-agent ☤ | | ------------------- | ------------------------------ | ---------------- | | Binary | 15 MB static | Python venv + uv | | Startup | < 50 ms | ~1–3 s | | Memory | ~15 MB resident | ~80–150 MB | | LLM providers | 14 built-in (+ Azure, Bedrock) | varies | | Messaging platforms | 15 gateways | 7 platforms | | Tests | 1629 passing (Rust) | — | | Migrate from hermes | edgecrab migrate | N/A |

---

Table of Contents

• EdgeCrab 🦀
  • Architecture
  • Table of Contents
  • Why EdgeCrab?
  • Quick Start (90 seconds)
    • Option A — npm (no Rust required)
    • Option B — pip (no Rust required)
    • Option C — cargo
    • Option D — build from source
    • Guided Setup Output
    • First Prompts
  • What EdgeCrab Can Do
    • ReAct Tool Loop
    • 74 Built-in Tools
      • File Tools (file toolset)
      • Terminal Tools (terminal toolset)
      • Web Tools (web toolset)
      • Browser Tools (browser toolset)
      • Memory & Honcho Tools (memory + honcho toolsets)
      • Skills Tools (skills toolset)
      • Session & Search (session toolset)
      • Delegation & MoA (delegation + moa toolsets)
      • Code Execution (code_execution toolset)
      • MCP Tools (mcp toolset)
      • Media Tools (vision / tts / transcribe toolsets)
      • Automation Tools
    • Sub-agent Delegation
    • Sandboxed Code Execution
    • Browser Automation
    • 15 Messaging Gateways
    • Persistent Memory & Learning
    • Skills Library
    • Cron Scheduling
    • Checkpoints & Rollback
    • Profiles & Worktrees
    • Vision, TTS & Transcription
  • 14 LLM Providers
  • 6 Terminal Backends
  • MCP Server Integration
  • ACP / VS Code Copilot Integration
  • ratatui TUI
  • All CLI Commands
  • All Slash Commands
  • Security Model
  • Architecture
  • Configuration
  • SDK: Python & Node.js
    • Python SDK (edgecrab-sdk)
    • Node.js SDK (edgecrab-sdk)
  • Docker
  • Migrating from hermes-agent
  • Testing
  • Project Structure
  • Requirements & Build
  • Contributing
  • Release Channels
  • License

---

Why EdgeCrab?

Most AI agents are either too constrained (coding agents that forget you exist after the session) or too heavy (Python runtimes, Node daemons, GBs of RAM). EdgeCrab is different.

It learns. Like Nous Hermes Agent, EdgeCrab maintains persistent memory across sessions, auto-generates reusable skills, and builds a cross-session Honcho user model that gets smarter over time.

It's everywhere. Like OpenClaw, EdgeCrab lives in your channels — Telegram, Discord, Slack, WhatsApp, Signal, Matrix, Mattermost, DingTalk, SMS, Email, Home Assistant, and more. Send it a voice memo on WhatsApp and get a PR back.

It's fast and lean. Unlike Python agents, EdgeCrab is a Rust binary. It starts before you finish blinking. It uses ~15 MB of RAM instead of 150 MB. Security is compiled in — path jails, SSRF guards, command scanners — not runtime patches.

It's extensible. MCP servers, custom Rust tools, Python/JS sandboxes, sub-agents, Mixture-of-Agents consensus — the full toolkit for heavy-duty automation.

---

Quick Start (90 seconds)

Option A — npm (no Rust required)

npm install -g edgecrab-cli
edgecrab setup               # interactive wizard — detects API keys, writes config
edgecrab doctor              # verify health
edgecrab                     # launch TUI

Option B — pip (no Rust required)

pip install edgecrab-cli
# OR: pipx install edgecrab-cli  (isolated install)
edgecrab setup && edgecrab doctor && edgecrab

Option C — cargo

cargo install edgecrab-cli
edgecrab setup && edgecrab doctor && edgecrab

Option D — build from source

git clone https://github.com/raphaelmansuy/edgecrab
cd edgecrab
cargo build --release         # ~30 s first build
./target/release/edgecrab setup

Guided Setup Output

EdgeCrab Setup Wizard
──────────────────────────────────────────────────────────────
✓ Detected GitHub Copilot (GITHUB_TOKEN)
✓ Detected OpenAI (OPENAI_API_KEY)

Choose LLM provider:
  [1] copilot      (GitHub Copilot — gpt-4.1-mini)  ← auto-detected
  [2] openai       (OpenAI — GPT-4.1, GPT-5, o3/o4)
  [3] anthropic    (Anthropic — Claude Opus 4.6)
  [4] ollama       (local — llama3.3)
  ...
Provider [1]: 1

✓ Config written to ~/.edgecrab/config.yaml
✓ Created ~/.edgecrab/memories/
✓ Created ~/.edgecrab/skills/

Run `edgecrab` to start chatting!

First Prompts

edgecrab "summarise the git log for today and open PRs"
edgecrab --model anthropic/claude-opus-4-6 "review this codebase for security issues"
edgecrab --model ollama/llama3.3 "explain this code offline"
edgecrab --quiet "count lines in src/**/*.rs"   # pipe-safe, no banner
edgecrab -C "continue-my-refactor"              # resume named session
edgecrab -w "explore that perf idea"            # isolated git worktree

---

What EdgeCrab Can Do

EdgeCrab is an autonomous agent. Give it a goal in natural language; it reasons, calls tools, observes results, and loops until the task is done. Here's what it can actually reach.

ReAct Tool Loop

EdgeCrab uses a Reason → Act → Observe loop (ReAct pattern) implemented in crates/edgecrab-core/src/conversation.rs. Each turn:

1. System prompt built once per session (SOUL.md, AGENTS.md, memories, skills, date/time, cwd) — cached for Anthropic prompt cache hits
2. LLM decides what to do next (including parallel tool calls)
3. Security check runs before every tool execution (path jail, SSRF guard, command scan)
4. Tool executes — file I/O, shell, web, code, sub-agents, browser, etc.
5. Result injected back into context
6. Loop until no more tool calls (task done), Ctrl-C, or 90-iteration budget exhausted
7. Context compression fires at 50% of context window — prunes old tool outputs, then LLM-summarizes
8. Learning reflection auto-fires after ≥5 tool calls — agent can save new skills and update memory

The budget default is 90 iterations (max_iterations in config). Increase it for long autonomous tasks.

74 Built-in Tools

Tools are registered at compile time via the inventory crate — zero startup cost. The ToolRegistry dispatches by exact name with fuzzy (Levenshtein ≤3) fallback suggestions.

File Tools (file toolset)

| Tool | What it does | | -------------- | ---------------------------------------------------------------------------- | | read_file | Read file with optional start_line/end_line — path-jailed, canonicalized | | write_file | Write or create file (parent dirs auto-created) | | patch_file | Search-and-replace patch — exact string match, atomic write