Menu
Start the A-Team Codex coordinator workflow for designing or restructuring teams
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Canonical validation checklist for generated team structures, executed by a fresh-context verifier
Entry point that makes this session adopt the Team Architect workflow to run the full team design process
Rewrite generated prompts to be sharper, shorter, and more executable without changing meaning
Apply Codex-native prompt patterns for agents, skills, rules, and project instructions
Validate that a generated Codex team is structurally complete and internally consistent
Provide systematic prompt optimization methodology to improve instruction quality while preserving original characteristics
Based on SOC occupation classification
| name | A-Team |
| description | Start the A-Team Codex coordinator workflow for designing or restructuring teams |
Use this skill when the user asks A-Team to design, generate, review, or restructure an agent team. This is the Codex-native entry skill; it complements AGENTS.md instead of replacing it.
$a-team, A-Team, or equivalent wordingAGENTS.md is the project-level runtime contract..codex/agents/team-architect.md is the coordinator playbook..codex/config.toml registers specialist subagents when multi-agent delegation is available.teams/{team-name}/..claude/ remains legacy/source design and must not be modified unless explicitly requested.Use this skill when the request matches one of these intents:
teams/{team-name}/Do not use this skill for unrelated coding tasks inside this repository.
AGENTS.md..codex/agents/team-architect.md.Treat the arguments as the initial team brief. Extract objective, target team name, requested format, and existing team path when present.
Ask one focused discovery question: what team the user wants A-Team to design or restructure.
.claude/ unless the user explicitly asks for Claude conversion or legacy updates.allowed-tools, disable-model-invocation, or context: fork in Codex-native skills..codex/agents/; use project-root agents/.Input: $a-team design a research writing team for long-form technical articles
Output: Start Discovery, confirm requested format and execution mode, then produce the requirements summary before planning.
Input: $a-team restructure teams/content-studio because review handoffs are too slow
Output: Run Phase 7, inspect the existing team, classify the handoff problem, and recommend the smallest effective structural change.
Input: $a-team
Output: Ask what team the user wants designed or restructured, then continue Discovery in the user's language.