- Home
- Skills
- Agents & Orchestration
- autocoding team
autocoding team
Deploy an autonomous two-agent team with a supervisor-worker hierarchy, Git worktrees, and scheduled systemd execution.
Free
Works with the AI Agents you already use
autocoding team
Example session with this skill installed
Set up a new autocoding team in ./projects/auth-refactor based on the plan in docs/migration-plan.md. Use Codex as the developer.
- Read your context and instructions
- Compiled the autocoding team
I have analyzed the prerequisites and the migration plan. The team is now configured to handle the authentication refactor using the auth-ref slug.
Team Scaffolding Results
| Component | Status | Path/Details |
|---|---|---|
| Team Folder | Created | ./projects/auth-refactor |
| Orchestrator | Initialized | orchestrator.md (Role: Supervisor) |
| Developer | Ready | developer_onboarding.md (Agent: Codex) |
| Worktree | Active | ../repo-auth-ref on branch team/auth-ref |
| Timer | Configured | auth-ref-dev.timer (5m interval) |
| Isolation | Verified | Private test environment in ensure_env |
Verification Logs
bash -nchecks passed for all maintenance scripts.systemd-analyzeverified unit files for the user session.- Smoke test of Codex CLI successful (Model: default, Effort: medium).
- No placeholder tags
{{...}}remain in the operational files.
Next steps
- Run
systemctl --user enable --now auth-ref-dev.timerin your terminal to start the developer. - Start the orchestrator session here by typing:
/loop Orchestrator tick: follow ./projects/auth-refactor/orchestrator.md §7. - Monitor
progress.txtfor real-time status updates on the migration phases.
Connects securely to your tools. The creator never sees your data.
About this skill
Setting up a multi-agent workflow often leads to orchestration overhead, conflicting instructions, and lack of oversight. This skill automates the deployment of a structured two-agent team—an Orchestrator and an unattended Developer—managed through a formal board and report system.
What it does
- Team Initialization creates a dedicated team folder with standardized
orchestrator.mdandboard.mdfiles based on your task and plan. - Environment Isolation provisions a Git worktree to separate the autonomous developer's workspace from your main checkout.
- Automated Scheduling configures
systemdtimers to run the developer agent at regular intervals without manual triggers. - Quality Gates enforces a supervisor-worker relationship where the developer submits reports for orchestrator review before progressing.
- Conflict Detection audits existing systemd units and Git worktrees to prevent overlaps with other active agent teams.
How it works
- Prerequisite Check scans for a
PREREQUISITES.mdcontaining your task description and implementation plan. - Configuration Intake asks a single batch of questions to confirm authority, spending limits, and test isolation strategies.
- Scaffolding generates the customized instructions, scripts, and systemd units needed for the specific project slug.
- Validation runs a smoke test of the agent CLI and verifies script syntax and systemd unit integrity.
Frameworks & tools
This skill works with the Codex CLI or Claude Code as the developer agent. It requires Git for worktree management and systemd (Linux) for task scheduling.
Why this beats prompting it yourself
Manual prompting for long-running autonomous tasks often fails due to context drift or lack of persistent state. This skill builds a durable operational framework with a decision log and status board, ensuring the agents follow a proven "run book" rather than improvising their workflow.
Use cases
- Long-running Refactors where an agent needs to work through a multi-phase plan over several hours.
- Parallel Feature Development by spinning up a secondary agent team in a separate worktree.
- Continuous Maintenance of documentation or tests that run on a timer.
Known limitations
Requires a Linux environment with systemd for the autonomous timer functionality. The owner must manually fill the prerequisites form before setup can complete.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Agents & Orchestration.
Requires Linux with systemd for timers. Requires Codex CLI or Claude Code. Main checkout must be a Git repo. Owner must manually provide a task description and plan.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 10 days ago
- Free with an Agensi account
- Software
Git