Agent orchestration handoff

Reusable AI agent skills, project templates, and guardrails for safer software maintenance and delivery.

Install

npx -y skills add jukrap/ai-agent-playbook --skill agent-orchestration-handoff

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

Copied from the file, not written here

Use when splitting work across agents, subagents, workers, review passes, or long-running handoffs.

1.4 KB, as published. Nobody here has run it

Use this as the primary AI harness skill for dividing multi-agent work, bounding worker context, recording evidence, and reconciling handoffs.

Classify the orchestration mode: parallel subtasks, sequential handoff, review pass, external worker, long-running resume, or reconciliation after multiple outputs.
Define a worker contract before dispatch: goal, allowed paths, required reads, forbidden writes, tool limits, expected artifacts, stop conditions, and context budget.
Keep a compact evidence ledger for each worker with task id, input scope, claims, locators, status, residual risk, and review result.
Reconcile outputs before merging conclusions: conflicts, duplicate edits, stale facts, missing verification, privacy leaks, and generated evidence boundaries.
Promote only reviewed conclusions into memory, docs, worklogs, or canon. Keep raw traces and generated reports under runtime or local-only evidence.

Read references/worker-contract.md when defining worker scope, allowed context, outputs, and stop conditions.

Read references/evidence-ledger-and-reconciliation.md when merging multiple worker outputs, reviewing claims, or preparing a handoff.