Guide accessibility reviews for UI, documents, and workflows. Use for WCAG-oriented audits, keyboard review, semantics, contrast, and remediation planning.
日本語の概要は準備中です。原文の説明を表示しています。
Orchestrate Kilo headless boss-worker execution with repo-local state, watchdog leases, consensus validation, and safety gates. Use when running multi-agent autonomous loops that require planning, delegation, verifier checks, and high/critical discrepancy adjudication.
日本語の概要は準備中です。原文の説明を表示しています。
実行スコア 未検証
インストール方法を見るインストールする前に、エージェントに与えられる指示の中身を確認できます。
Purpose: run a durable boss-worker gentle loop for headless Kilo execution with explicit state, safety controls, and consensus validation.
Use this skill when you need:
session_id.command_argv lists beginning with kilo run; shell command strings are rejected.gate_passed: true and requires_tiebreaker: false..kilo/state/orchestrator/..kilo/state/orchestrator/queue.jsonl.kilo/state/orchestrator/tasks/.kilo/state/orchestrator/results/.kilo/state/orchestrator/leases/.kilo/state/orchestrator/heartbeats/.kilo/state/orchestrator/sessions/.kilo/state/orchestrator/consensus/.kilo/state/orchestrator/deadletter.jsonl.kilo/state/orchestrator/events.logscripts/kilo_headless_runner.py: execute a single worker card through kilo run.scripts/boss_ctl.py: queue, dispatch, status, consensus, watchdog, and completion flows.scripts/kilo_retry_watchdog.py: deterministic phase-1 recovery helper that detects failure markers, sends fixed resume prompts to tmux on 15s cadence (configurable), and emits Hermes outbox JSON on retry exhaustion.Use when orchestration has transient failure and no AI fallback/backstop is available:
This script is intentionally offline-capable and non-AI.
python3 scripts/boss_ctl.py init
python3 scripts/boss_ctl.py enqueue --task-file references/examples/task.sample.yaml
python3 scripts/boss_ctl.py dispatch --max-dispatch 2
python3 scripts/boss_ctl.py status
python3 scripts/boss_ctl.py watchdog
python3 scripts/boss_ctl.py consensus --task-id <task-id>
python3 scripts/boss_ctl.py finalize --task-id <task-id>
Recommended trigger strategy:
Use ls-cron-job-scheduling with a user-created manifest, commonly cron/manifest.yaml in the target repo.
references/kilo-boss-orchestrator-runbook.mdreferences/kilo-boss-state-schema.mdreferences/kilo-headless-boss-worker-gentle-loop-architecture.mdreferences/examples/task.sample.yamlまだレビューはありません。使ってみた感想をお寄せください。
ログインなしで書き込めます。使ってみた感想、質問、動かなかった環境などを気軽にどうぞ。
まだコメントはありません。最初のコメントを書いてみませんか。
Guide accessibility reviews for UI, documents, and workflows. Use for WCAG-oriented audits, keyboard review, semantics, contrast, and remediation planning.
日本語の概要は準備中です。原文の説明を表示しています。
Use when choosing a LocalSetup Agent-* lane for exploration, research, bounded implementation, or review. Consult the static capability matrix; do not probe account or network availability.
日本語の概要は準備中です。原文の説明を表示しています。
Guide agent email inbox workflows with prompt-injection defenses. Use for inbound email processing, webhooks, routing, and safety review.
日本語の概要は準備中です。原文の説明を表示しています。
Navigate and understand codebases using agentlens hierarchical documentation. Use when exploring new projects, finding modules, locating symbols in large files, finding TODOs/warnings, or understanding code structure.
日本語の概要は準備中です。原文の説明を表示しています。
Use for Agent Q signed encrypted file-drop or mail ship and ingest, private registry v2 setup, queue status, and transport maintenance.
日本語の概要は準備中です。原文の説明を表示しています。
Inventory wrapper for the alirezarezvani Claude skills bundle. Use when triaging its broad skill catalog and large script surface before any targeted import.
日本語の概要は準備中です。原文の説明を表示しています。