实验室Lab / Grok Bot / gawkbot:开源、本机跑的 Grok Bot 替代 gawkbot: open-source, on-machine Grok Bot alternative

社区 Community 需审批 Needs approval 例行 Routine 多Agent Multi-agent

gawkbot:开源、本机跑的 Grok Bot 替代 gawkbot: open-source, on-machine Grok Bot alternative

gawkbot: open-source, on-machine Grok Bot alternative

本机多 Bot(各自目录与工具白名单)、外发/提交/采购/删除逐次审批、可接 Claude Code / Codex / 本地模型,并给结果做 microapp 看板。对比表写明相对 xAI Grok Bot 的免费与本地差异。非 xAI 产品。 Local multi-bots (per-bot directories and tool allowlists), per-action approval for send/commit/purchase/delete, Claude Code / Codex / local models, plus microapp dashboards for outcomes. Comparison table vs xAI Grok Bot (free, on-machine). Not an xAI product.

这条用法在做什么What this use does

本机多 Bot(各自目录与工具白名单)、外发/提交/采购/删除逐次审批、可接 Claude Code / Codex / 本地模型,并给结果做 microapp 看板。对比表写明相对 xAI Grok Bot 的免费与本地差异。非 xAI 产品。 Local multi-bots (per-bot directories and tool allowlists), per-action approval for send/commit/purchase/delete, Claude Code / Codex / local models, plus microapp dashboards for outcomes. Comparison table vs xAI Grok Bot (free, on-machine). Not an xAI product.

收录理由Why we listed it Show HN item 49531991(分低)。GitHub ~1253★(检索日)。本机凭据风险自负;收录不等于安全审计。 Show HN item 49531991 (low points). GitHub ~1253★ on lookup day. Local credentials are your risk; listing is not a security audit.

上手路径How to start

  1. 先跑一次安全范围Run a safe first pass

    先在可撤回、不外发的范围内试跑一遍,确认输出可用,再考虑扩大权限。 Try it first in a reversible, non-outbound scope. Confirm the output, then widen permissions.

  2. 存 SkillSave a Skill

    把跑通的步骤存成 Skill,下次同类任务直接复用,而不是重新口述。 Save the working steps as a Skill so the next similar job reuses them instead of being re-explained.

  3. 再上 RoutineThen add a Routine

    输出稳定后再打开例行 / 定时。本条已有例行形态,不要一上来就无人值守外发。 Turn on the routine / schedule only after output is stable. This item already has a routine shape — do not start unattended outbound.

相关案例Related cases