kanban-worker

Guides Hermes Kanban workers through handoffs, retries, and edge cases.

1|Updated Jun 19, 2026
One-click install
npx skills add https://github.com/Lento47/arcana-community --skill kanban-worker-lento47
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: kanban-worker
Source: https://github.com/Lento47/arcana-community/tree/main/skills/devops/kanban-worker
Command: npx skills add https://github.com/Lento47/arcana-community --skill kanban-worker-lento47

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Dispatched Kanban workers often fail silently by producing vague block reasons, hallucinating created card IDs, skipping heartbeats, or mishandling shared workspaces. This Skill provides the deeper operational detail beyond the auto-injected lifecycle guidance so workers complete, block, and hand off tasks correctly. ## Core Features & Use Cases - Structured handoffs: Patterns for kanban_complete summaries and metadata shaped for downstream parsers, including review-required flows using kanban_comment plus kanban_block. - Workspace and tenant handling: Rules for scratch, shared dir:, and git worktree workspaces, plus tenant-prefixed memory entries to prevent cross-tenant leakage. - Retry diagnostics: Interprets prior run outcomes (timed_out, crashed, spawn_failed, reclaimed, blocked) so retries avoid repeating failed paths. - Use Case: A worker dispatched to fix a rate limiter finishes with 14 passing tests, posts structured metadata as a comment, and blocks with a review-required: reason so a human reviewer can approve before merge. ## Quick Start Load the kanban-worker skill and follow its handoff patterns to complete your current Kanban task with a structured summary and metadata.

Frequently Asked Questions about kanban-worker

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

FAQPage Schema
How do I complete a Kanban task as a Hermes worker agent?▼

Call kanban_complete with a concise summary and structured metadata such as changed files, test counts, and decisions. If you created follow-up cards, pass their captured IDs in created_cards so the kernel can verify them.

How should a Kanban worker ask a human for a decision?▼

Post full context with kanban_comment, then call kanban_block with a one-sentence reason naming the specific decision needed. Never call clarify, since headless workers have no live user and the call will time out silently.

What happens if I claim Kanban card IDs I did not create?▼

The kernel verifies each ID in created_cards against cards actually created by your profile. Phantom or hallucinated IDs cause the completion to be rejected, and the failed attempt is permanently recorded on the task's event log.

How do I handle a retry of a previously failed Kanban task?▼

Run kanban_show first and inspect prior runs' outcome, summary, and error fields. Outcomes like timed_out, crashed, or spawn_failed each indicate different fixes, such as chunking work, reducing memory, or blocking to ask a human.

When should a worker use kanban_block instead of kanban_complete?▼

Block when work needs human review or a decision, prefixing the reason with 'review-required: ' for code changes awaiting approval. Use kanban_complete only for genuinely terminal tasks like typo fixes or finished research writeups.