Stats
Actions
Tags
From relay
Execute a pre-written relay plan unsupervised using multi-agent workflows. Use when implementing a plan that was shaped for parallel, fan-out execution (independent stand-alone phases). The plan file path is passed as arguments.
How this skill is triggered — by the user, by Claude, or both
Slash command
/relay:exec-workflowThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Implement the plan at the path given in the arguments, using the Workflow tool to fan
Implement the plan at the path given in the arguments, using the Workflow tool to fan the work out across parallel agents. The plan was written as independent, stand-alone phases for exactly this — orchestrate them as a workflow. (Invoking this skill is the user's explicit opt-in to multi-agent orchestration.)
The user will be unavailable during execution — do not ask clarifying questions and do not stop until the plan is completely fulfilled.
Keep the build green and tests passing as you go. When you're done, review your own work before you call it finished — re-read the diff against the plan, check you didn't miss anything or leave something half-done. Don't commit anything.
If you encounter a decision the plan didn't cover, or a step that doesn't work against the real code, make the call yourself and keep going. Note what you decided or changed in a "decisions and deviations" section at the end.
Plan file: $ARGUMENTS
npx claudepluginhub danielpivonka/claude-relay-skills --plugin relayGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.