Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/TheBeardedBearSAS/claude-craftWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/thebeardedbearsas/claude-craft/sub-agents-patterns)<a href="https://agentmods.dev/commands/thebeardedbearsas/claude-craft/sub-agents-patterns"><img src="https://agentmods.dev/badge/commands/thebeardedbearsas/claude-craft/sub-agents-patterns/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/thebeardedbearsas/claude-craft/sub-agents-patterns"><img src="https://agentmods.dev/badge/commands/thebeardedbearsas/claude-craft/sub-agents-patterns.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00012 | $0.02793 |
| Opus 5.5 | $0.00005 | $0.01117 |
| Sonnet 5.5 | $0.00002 | $0.00559 |
| Haiku 4.5 | $0.00001 | $0.00279 |
Grade A, and why
sub-agents-patterns scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 278 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Patterns de Sub-Agents
Guide pour utiliser efficacement les sub-agents dans Claude Code pour les tâches parallèles et complexes.
Types d'Agents
1. Explore Agent (Recherche Rapide)
À utiliser pour l'exploration rapide du codebase et la collecte d'informations.
Task tool avec subagent_type: "Explore"
- Recherches rapides de patterns de fichiers
- Recherches de mots-clés dans le code
- Compréhension de la structure du codebase
Quand l'utiliser :
- Recherche de fichiers par pattern
- Recherche de patterns de code spécifiques
- Réponses aux questions sur l'organisation du codebase
2. General-Purpose Agent (Tâches Complexes)
À utiliser pour les tâches multi-étapes nécessitant de l'autonomie.
Task tool avec subagent_type: "general-purpose"
- Refactoring complexe
- Mises à jour multi-fichiers
- Recherche et implémentation
Quand l'utiliser :
- Tâches couvrant plusieurs fichiers
- Sous-tâches indépendantes pouvant s'exécuter en parallèle
- Tâches nécessitant du jugement et de l'itération
3. Plan Agent (Architecture)
À utiliser pour concevoir des stratégies d'implémentation.
Task tool avec subagent_type: "Plan"
- Planification d'implémentation
- Décisions d'architecture
- Analyse des compromis
Quand l'utiliser :
- Avant d'implémenter des fonctionnalités complexes
- Lorsque plusieurs approches sont possibles
- Pour les décisions architecturales
Patterns de Tâches Parallèles
Pattern 1 : Recherche Parallèle
Lancez plusieurs Explore agents pour différents aspects :
# Lancement en parallèle (un seul message avec plusieurs appels d'outils) :
- Agent 1 : Rechercher les patterns d'authentification
- Agent 2 : Rechercher les endpoints API
- Agent 3 : Rechercher les modèles de base de données
Pattern 2 : Mises à Jour Parallèles
Pour les mises à jour de fichiers indépendants entre langues/modules :
# Lancement en parallèle :
- Agent 1 : Mettre à jour les templates français
- Agent 2 : Mettre à jour les templates espagnols
- Agent 3 : Mettre à jour les templates allemands
- Agent 4 : Mettre à jour les templates portugais
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 278 lines · 12 tokens per session scan A 7b141ea602ea
sub-agents-patterns is a command published in the GitHub repository TheBeardedBearSAS/claude-craft (107 stars, last pushed 7d ago), licensed MIT. It adds 12 tokens to every session and 2,793 once invoked, about $0.0000 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other commands, from other repositories
assemble-team
Assemble a pre-built agent team for parallel work - review, feature, debug, cross-platform, full-stack, or research.
auto-claude
Invoke Auto-Claude autonomous coding framework for complex feature implementation.
handoff
Write a session handoff document for continuity.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.