todero

Interact with the Todero control plane API for agent task coordination and governance.

Updated Mar 26, 2026
One-click install
npx skills add https://github.com/nabitllc/todero --skill todero-nabitllc
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: todero
Source: https://github.com/nabitllc/todero/tree/main/skills-releases/todero/v0
Command: npx skills add https://github.com/nabitllc/todero --skill todero-nabitllc

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? AI agents running inside Todero need a disciplined way to receive work, execute it, and report back without human babysitting. This Skill teaches an agent the complete heartbeat procedure: authenticate with injected environment variables, check assignments, check out issues, do the work, update status, delegate subtasks, and handle approvals, blockers, and review gates through the Todero REST API. ## Core Features & Use Cases - Heartbeat execution loop: Follow a nine-step procedure covering identity lookup, inbox triage, issue checkout with conflict handling, context loading, work execution, and final status disposition. - Governance and approvals: Request board approvals, respond to execution-policy review stages, handle MCP tool approval gates, and create structured issue-thread interactions such as confirmations, checkbox selections, and item verdicts. - Dependency and delegation management: Set first-class blockers with blockedByIssueIds, create child issues with workspace inheritance, manage recurring routines, and upload artifacts as work products. - Use Case: An engineer agent wakes on a comment-triggered heartbeat, reads the wake payload, checks out the issue, fixes the reported bug, uploads a walkthrough video as an artifact, and patches the issue to in_review with a markdown comment linking the related approval. ## Quick Start Ask the agent to check its Todero inbox, pick up the highest-priority assigned issue, and complete it following the heartbeat procedure.

Frequently Asked Questions about todero

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I check out and update a Todero issue from an agent?

POST to /api/issues/{issueId}/checkout with your agentId and expectedStatuses, including the X-Todero-Run-Id header. After completing work, PATCH /api/issues/{issueId} with a new status such as done or in_review plus a comment explaining the outcome.

How does the Todero heartbeat procedure work for agents?

Each heartbeat follows nine steps: get identity, handle approval follow-ups, fetch assignments via inbox-lite, pick work by priority, check out the issue, load context, do the work, update status with a final disposition, and delegate subtasks if needed. Scoped wake payloads let you skip straight to checkout.

What should an agent do when a Todero issue checkout returns 409?

A 409 Conflict means another agent owns the task. Never retry it. Stop working on that issue, pick a different assigned task from your inbox, or exit the heartbeat if nothing else is assigned.

How do I request board approval in Todero?

POST to /api/companies/{companyId}/approvals with type request_board_approval, your agent id, linked issueIds, and a payload containing title, summary, recommended action, and risks. Todero wakes you with the approval decision via PAPERCLIP_APPROVAL_ID.

Can agents create recurring scheduled tasks in Todero?

Yes, agents can create routines assigned to themselves via POST /api/companies/{companyId}/routines with schedule, webhook, or API triggers. Each firing creates an execution issue the agent picks up in its normal heartbeat flow.

Why must agents include the X-Todero-Run-Id header on API requests?

The run ID header links every issue mutation (checkout, update, comment, subtask creation) to the current heartbeat run for audit traceability. Omitting it breaks the run audit trail that Todero uses to track which execution performed each action.