WorkflowsMay 14, 2026·2 min read

SafestClaw — Offline-First Assistant CLI

Install SafestClaw with pipx to run an offline-first assistant CLI (LLM optional) with commands for news, crawling, reminders, and more.

Agent Toolkit
Agent Toolkit · Community
Agent ready

This asset can be read and installed directly by agents

TokRepo exposes a universal CLI command, install contract, metadata JSON, adapter-aware plan, and raw content links so agents can judge fit, risk, and next actions.

Stage only · 17/100Stage only
Agent surface
Any MCP/CLI agent
Kind
CLI Tool
Install
Stage only
Trust
Trust: Established
Entrypoint
Asset
Universal CLI install command
npx tokrepo install f8dd69d2-5e10-56da-b555-773202d6e9e5
install contractmetadata JSONadapter planraw content
Intro

Use SafestClaw when you want an assistant-style CLI that can run without mandatory LLM APIs and still support useful local workflows.

Install SafestClaw with pipx to run an offline-first assistant CLI (LLM optional) with commands for news, crawling, reminders, and more.

Best for: developers who want a local assistant CLI with optional LLM usage and a focus on safer, command-driven workflows

Works with: Python 3.11+, pipx or venv, optional extras for MCP/CalDAV as documented in the README

Setup time: ~6–12 minutes (pipx + first run)

Key facts (verified)

  • GitHub: 279 stars · 35 forks · pushed 2026-05-09.
  • License: MIT · owner avatar + repo URL verified via GitHub API.
  • Entrypoint (README-backed): pipx install safestclaw.

Main

  • Prefer explicit commands (news/crawl/remind/calendar) so runs are reproducible and easier to audit than free-form chat.
  • If you enable optional integrations (email/calendar), start with read-only scopes and keep credentials isolated.
  • For team usage, document which extras are enabled and which external services are allowed.

README excerpt (source)

# SafestClaw 🐾

**The zero-cost alternative to OpenClaw. No LLM required, though it is optional. No required API bills. Minimal attack surface. Runs on any machine.**

FAQ

Q: Do I need an LLM API key? A: The README describes LLM as optional; start without it and enable only if needed.

Q: What’s a safe first workflow? A: Try news/crawl/summarize on public URLs before connecting private accounts.

Q: How do I reduce risk? A: Keep integrations minimal and prefer whitelisted sources for crawling and feeds.

🙏

Source & Thanks

Source: https://github.com/princezuda/safestclaw > License: MIT > GitHub: 279 stars · 35 forks · pushed 2026-05-09

Discussion

Sign in to join the discussion.
No comments yet. Be the first to share your thoughts.

Related Assets

Claudable — Web Builder Powered by Local CLI Agents

Claudable is an open-source web builder that leverages local CLI agents (Claude Code, Codex, Gemini) with a repo-first quick start (3,951★).

Workflows
AI Open Source

Bernstein — Audit-Grade Orchestrator for CLI Agents

Bernstein coordinates CLI coding agents in parallel worktrees with signed audit chains, deterministic scheduling, and evidence trails.

Workflows
Agent Toolkit

MemPalace — Local-First AI Memory CLI

MemPalace is a local-first memory system with a `mempalace` CLI (+ MCP tools), hitting 96.6% R@5 on LongMemEval with no LLM required.

CLI Tools
Agent Toolkit

Bifrost CLI — Run Claude Code with Any AI Model

Enterprise AI gateway that lets Claude Code use any LLM provider. Bifrost routes requests to OpenAI, Gemini, Bedrock, Groq, and 20+ providers with automatic failover.

Workflows
AI Open Source
◈Home🔍Search👤Me