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

Squad Builder

ASecurity

Use when building an AI agent squad from scratch through owner interviews, external research, role design, and gated rollout. Triggers on \"squad builder\", \"agent team from scratch\", \"open squad\", \"discover interview agents\", \"five roles squad\". Non-triggers: picking ready-made agents from a list (use team-builder). Outcome: a five-role squad with researched briefs, image and publish skills wired, and human sign-off recorded at every gate.

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

Security Analysis

A100/100

Scanned 9/19/2026

Install to Claude Code

$npx -y skills add majinmagros/magros.ai-skills --skill squad-builder --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Squad Builder?

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

Security grade badge for Squad Builder
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/majinmagros-squad-builder/badge)](https://www.skillsdirectory.com/skills/majinmagros-squad-builder)

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

Download Zip
Files
SKILL.md
---
name: squad-builder
description: "Use when building an AI agent squad from scratch through owner interviews, external research, role design, and gated rollout. Triggers on \"squad builder\", \"agent team from scratch\", \"open squad\", \"discover interview agents\", \"five roles squad\". Non-triggers: picking ready-made agents from a list (use team-builder). Outcome: a five-role squad with researched briefs, image and publish skills wired, and human sign-off recorded at every gate."
metadata:
  origin: ECC
---

# Squad Builder

Interviewer-led squad construction in four gated stages: discover the
owner, investigate the outside world, design the squad, then run it.
`team-builder` composes ready-made agents from a menu; this skill
creates the squad from zero and refuses to skip the human gates.

## When To Activate

- The user says agent team from scratch, open squad, build my squad,
  interview me, or five agent roles.
- No agent roster exists yet and the squad must match a real owner,
  niche, and channel mix.
- A previous auto-generated team felt generic and missed the voice,
  offers, and constraints of the business.
- Image and publishing steps (screenshots, carousels, posting) must be
  part of the squad from day one.

## Stage Gates

```text
discover -> GATE 1 -> investigate -> GATE 2 -> design -> GATE 3 -> run
```

- Every arrow is a mandatory human gate. No gate is implied, batched,
  or auto-approved.
- Each gate records: approved, approved-with-changes, or rejected with
  reason. Rejected stages restart, they never slide forward.
- No agent is spawned before GATE 3 passes.

## Workflow

### 1. Discover - interview the owner

- Run a short interview, one decision-forcing question at a time in
  the style of `grilling`: niche, offers, audience, voice, channels,
  constraints, weekly capacity, approval appetite.
- Capture answers verbatim first, then compress into a one-page owner
  brief: goals, non-goals, risks, done criteria.
- End with GATE 1: the owner signs the brief or corrects it. No
  research starts on an unsigned brief.

### 2. Investigate - research outside the room

- Assign one researcher role (call it Sherlock): profile the niche on
  public IG and YT channels, video download helpers where allowed,
  plus repo and docs search for comparable squads.
- Deliver an evidence pack per claim: source, date, what was observed,
  what it implies for the squad. Unsourced claims are marked as such.
- Cover at minimum: top formats in the niche, posting cadence that
  sustains, hooks that repeat, and failure shapes to avoid.
- End with GATE 2: the owner keeps, cuts, or redirects each finding.
  Cut findings leave the pack, they do not linger as context.

### 3. Design - five roles plus wiring

- Draft exactly five roles by default: strategist, researcher,
  writer, visual builder, publisher. Rename to the niche, keep the

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 →