本文へ移動
cccskills
無料GitHub で公開

pb-project-new

Start a new private GitHub project the AOS way: create the private repo, bootstrap AOS (project card, AGENTS.md, wiki, memory, CI choice), push after GO, and confirm the first CI run is green and the repo is still private. Use for "new project", "start a repo", "bootstrap a project".

インストール方法を見る

含まれるファイル(1)

  • SKILL.md3.6 KB

SKILL.md(原文)

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。

New private AOS project

What you get: a private GitHub repo with AOS bootstrapped, pushed, and a first CI run checked.

Inputs

  • name, owner — asked in step 1
  • local_folder — where the clone goes, asked in step 1
  • upstream_url (optional) — a repo to track as upstream, asked in step 1

Steps

  1. Ask — name, owner, local folder, optional upstream URL → run log header in the start directory (production_artifacts/pb-project-new-<date>.md, never committed) — all answered
  2. [GO] github — from the parent of <local_folder> (cd there first; the clone lands in <name>/), gh repo create <owner>/<name> --private --clone → repo and local clone — The run stops here until the human types GO.
  3. github — gh repo view <owner>/<name> --json visibility -q .visibility → value in the run log — must equal PRIVATE; otherwise fail-stop: write the value to the run log, tell the user, push nothing, and do not delete or change the repo (destructive actions need the user's confirmation)
  4. Upstream given → git remote add upstream <url> — git remote -v shows it; never push to upstream (no gh repo fork: a fork of a public repo cannot be private)
  5. aos-project-init — folder → .aos/project.json, AGENTS.md and symlinks, OpenWiki, memB project card, Synapse, CI choice (it may create triage labels and write ../WORKTREE.md; aos-project-init asks for its own write-list approval) — doctor rows green or exceptions named, .openwiki/ present; accept aos-project-init's own commit if it makes one, step 6 commits only what is left, never the run log
  6. git commit chore: bootstrap AOS project — remaining bootstrap files → SHA — commit succeeds (any hook failure is reported, never bypassed with --no-verify) (skip if nothing is left)
  7. [GO] git branch -M main, then git push -u origin main — SHA → remote — The run stops here until the human types GO.
  8. pb-ci-fix — only if CI was chosen in step 5: run its step 10 (watch the run on the pushed SHA) → conclusion success; if red, hand over to /pb-ci-fix from its step 1 (it has its own GO); then re-run the step 3 check — still PRIVATE

Run log: production_artifacts/pb-project-new-<date>.md in the start directory, outside the new repo

Rules

  • Anything other than the literal GO (case-insensitive) is not a GO; a GO covers only that one step, one time.
  • GO is by contract in every harness. The go-gate hook is only a backstop on Claude Code, OpenCode and agy (Codex: unverified; Cursor, Kimi: none). A missing hook is never permission to proceed.
  • A failed check stops the run: write the failure into the run log and report. No silent retries.
  • Write one run-log line per step as it completes (N. done|skipped|failed — artifact — check result) and WAITING FOR GO: <step> at each gate.

レビュー

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。

同じリポジトリのスキル

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。

hybridlabor-api/aos62026年10月12日 更新

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。

hybridlabor-api/aos62026年10月12日 更新

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。

hybridlabor-api/aos62026年10月12日 更新

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。

hybridlabor-api/aos62026年10月12日 更新

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。

hybridlabor-api/aos62026年10月12日 更新

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。

hybridlabor-api/aos62026年10月12日 更新

hybridlabor-api のスキルをすべて見る

このスキルの問題を報告する