Use when manage multiple local CLI agents via tmux sessions (start/stop/monitor/assign) with cron-friendly scheduling.
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 proactively checking for and installing updates to the AOS (BDB Agent OS) package via npm.
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。
Keeps AOS (@hybridlabor-api/aos) up to date. Since v4.0.0 (renamed from
@hybridlabor-api/bdb-dev-optimized-agent-skills, which is deprecated on
npm and frozen at its last pre-rename version — never check that name,
it will never report a real update again).
npm view @hybridlabor-api/aos dist-tags --json
Compare against the locally installed version (~/.agents/aos/package.json,
or wherever the primary install target's package.json lives — the
installer's own verifyEcosystemInstallation() status table is the
authoritative source if you're running from inside an existing install).
Two dist-tags exist:
@latest — the stable channel. CI (release-please) publishes here
automatically on every release; this is what real users should run.@next — an ad-hoc staging tag used occasionally to validate a large
change (a rename, a major version bump) before promotion to @latest.
It is not a standing parallel channel — don't default to it unless the
user is specifically testing a migration.npx -y @hybridlabor-api/aos@latest
-y runs it fully non-interactively. If an install already exists, this
takes the Quick Update path (refreshes skills, re-syncs installed
submodules, refreshes the OpenWiki daemon schedule) rather than a full
reinstall — it does not need a --force flag or any special handling for
"already installed."
The installer syncs skills to every detected harness in one pass, not
just one: ~/.claude/skills, ~/.gemini/config/skills (Antigravity),
~/.agents/skills, ~/.codex/skills, ~/.cursor/skills, ~/.roo/skills.
Don't assume Gemini/Antigravity is the only target.
After running, verifyEcosystemInstallation()'s own status table (printed
at the end of the run) is the ground truth — it checks each tracked
submodule's actual installed version against its own npm dist-tags
(bdb-synapse, memB, heimdall-token-saver, bdb-dev-creator-extension,
bdb-os-remote, bdb-dev-tool-installer, and aos itself). Read that
table rather than assuming success from a clean exit code — a partial
failure (e.g. a daemon not responding on its port) still exits 0 and logs
a warning, not an error.
Report the new version and anything flagged as a warning in that table (not just "update complete"). Don't report success without having actually looked at the status table's contents.
If the user wants recurring checks, recommend /schedule with the current
package name:
/schedule CronExpression="0 10 * * 1" Prompt="Check if there is a new version of @hybridlabor-api/aos via npm view and update it"
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。
概要と使いどころ
Use when manage multiple local CLI agents via tmux sessions (start/stop/monitor/assign) with cron-friendly scheduling.
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 a hybrid memory system that provides persistent, searchable knowledge management for AI agents (Architecture, Patterns, Decisions).
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。
Reference for how BDB structures autonomous software engineering work — the seven-node dispatcher graph (Architect, TechLead, UI/UX, Engineering, Media/EventTech, Reviewer, Shipping) that /startcycle-graph actually runs. Use when you need the high-level lifecycle framing without inventing your own process.
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。
Tools are how AI agents interact with the world. A well-designed tool is the difference between an agent that works and one that hallucinates, fails silently, or costs 10x more tokens than necessary. This skill covers tool design from schema to error handling.
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。
Harness patterns for coding agents — memory, permissions, context engineering, delegation, skills, hooks, bootstrap.
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。
agenttrail: live map of a multi-agent build in the browser: which plan component is being worked on, by which agent or harness, what is done and what is stuck. Use when a multi-agent pipeline starts (/startcycle, /startcycle-graph, /teamwork-preview) or after a plan-canvas approve, when the user asks to see what the agents are doing ("live map", "build board", "who is running now"), or by yourself whenever a multi-agent run is under way — no user prompt needed.
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。