Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Grokbot Team Ops

ASecurity

Use when building teams of single-job bots with shared templates, dual memory, unified computer, teach-by-recording, routines, and visible handoffs. Triggers on \"bot team\", \"grokbot\", \"teach-a-task\", \"dual memory\", \"bot routines\", \"group chat bots\". Non-triggers: single human-plus-agent workspace (use buzz-workspace-teaming), one-off file task automation (use claude-cowork-patterns). Outcome: a bot team with roles, templates, memory policy, routines, approvals, and full logs.

2 stars
0 votes
0 copies
0 views
Added 9/19/2026
ai-agentsgoapi

Works with

terminalcliapi

Security Analysis

A100/100

Scanned 9/19/2026

Install to Claude Code

$npx -y skills add majinmagros/magros.ai-skills --skill grokbot-team-ops --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Grokbot Team Ops?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Grokbot Team Ops
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/majinmagros-grokbot-team-ops/badge)](https://www.skillsdirectory.com/skills/majinmagros-grokbot-team-ops)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
name: grokbot-team-ops
description: "Use when building teams of single-job bots with shared templates, dual memory, unified computer, teach-by-recording, routines, and visible handoffs. Triggers on \"bot team\", \"grokbot\", \"teach-a-task\", \"dual memory\", \"bot routines\", \"group chat bots\". Non-triggers: single human-plus-agent workspace (use buzz-workspace-teaming), one-off file task automation (use claude-cowork-patterns). Outcome: a bot team with roles, templates, memory policy, routines, approvals, and full logs."
metadata:
  origin: ECC
---

# Grokbot Team Ops

Run a factory of single-job bots: one bot per job, a boss bot that
orchestrates, bots that delegate bot-to-bot, shareable templates, and
visible handoffs in group chats. Vendor plans and prices change fast -
check the current vendor plan before acting; no price is recorded here.

## When To Activate

- The user says bot team, bot factory, boss-of-bots, or names single-job
  bots (miner, writer, scheduler, inbox triage).
- The task needs roles with fixed jobs plus delegation between bots.
- The task needs routines (time or event triggers) plus webhook doorbells.
- A previous multi-bot run lost memory, skipped approvals, or hid handoffs.

## Workflow

1. Define roles: one job per bot (single-job descriptions), plus one boss
   bot that owns sequencing and acceptance. Write each description so any
   bot can delegate to it without extra context.
2. Package templates: make each team shareable by link (roles, routines,
   plugins, prompts). Version templates; never fork silently.
3. Set dual memory: global memory (team facts, brand voice, policies) vs
   per-bot memory (job state, credentials scope, routine history). Per-bot
   memory never leaks into global without an explicit promote step.
4. Attach one computer: persistent browser logins, shared workspace,
   terminal, and local-file access per bot scope. Least privilege first.
5. Teach by recording: for click-heavy jobs, record the navigation once,
   convert to a skill (record-shadow-to-skill), then review the generated
   skill before Pocket deploy. Prefer API or connector when one exists.
6. Schedule routines: time triggers (daily summaries, 9:05 weekday digest)
   plus event triggers plus webhook doorbells (external ping -> bot wakes).
   Every routine declares output channel and done condition.
7. Organize chats: org chart (C-suite + workers), group chats with
   mentions, and visible handoffs (who passed what to whom, with link).
8. Gate risk: approvals and human-takeover for money, publishing, or
   irreversible actions. Log everything: prompts, tool calls, handoffs,
   approvals. No log = did not happen.

## Anti-Patterns

- Multi-job mega-bot that does mining, writing, and posting in one role.
- Shared memory soup with no global vs per-bot boundary.
- Routines with no done condition or no output channel.
- Silent handoffs (DM side-channels nobody can audit).
- Auto-approving money, publishing, or deletes to "go faster".

## Relations

- `buzz-workspace-teaming`: human-plus-agent workspace the bot team plugs
  into (channels, inbox, 24/7 hosting).
- `claude-cowork-patterns`: single-task execution patterns reused per bot.
- `routines`: scheduling primitive behind step 6 (time/event triggers).
- `criar-skill`: turns a recording into a reviewed, versioned skill.

## Sources

- No vendor docs URL confirmed at write time - verify the current vendor

## Exemplo

```text
Time: miner (coleta) + writer (redação) + boss (sequência e aceite)
Memória: global (voz, políticas) vs por-bot (estado do job) — promote explícito
Rotina 9h05: digest no canal + handoff visível; dinheiro/publicar só com approval
```

## Receita AgentMail: inbox proprio do bot via webhook (auditoria semanal 13-18/09)

Para dar ao bot email proprio: crie inbox em agentmail.to, conecte no marketplace, crie o bot, configure rotina webhook, copie URL + Authorization Bearer para os endpoints (custom headers) e filtre por inbox_id via edit-transformation JS (a UI nao tem dropdown). AVISO: URL+key equivalem a senha — trate como segredo. Protocolo de teste: envie email-teste, confira spinning apenas no bot certo e reply-in-thread antes de ligar a rotina. Ver routines para gatilhos schedule/webhook.

Attribution

majinmagrosmajinmagros
View sourceMore from majinmagros →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.

1023331 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

686011 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3331 votes

catchup

Recovers prior coding-agent session context by running `catchup <agent> --since-compact`, which extracts a clean summary of a previous Codex, Claude Code, Antigravity, OpenCode, or Pi Agent session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", or asks to recover/summarize a previous session before continuing. Do NOT use for the current conversation, git history, or any non-agent log.

611 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →