本文へ移動
cccskills
無料GitHub で公開

kelly-agent-builder

Busabase-backed App-in-Skill low-code agent configuration and governance console for a platform team. Use when the user invokes $kelly-agent-builder or /kelly-agent-builder, wants to review or edit a catalog of mock LLM agent configs, check quota usage, find configs that need attention, move a draft to live, pause a live agent, or archive an agent. Mock config/governance console only — it never provisions or calls any real agent.

インストール方法を見る

含まれるファイル(57)

  • SKILL.md8.3 KB
  • agents/openai.yaml268 B
  • assets/screenshots/agent-detail-zh-CN.webp130 B
  • assets/screenshots/agent-detail.webp130 B
  • assets/screenshots/catalog-zh-CN.webp130 B
  • assets/screenshots/catalog.webp130 B
  • assets/screenshots/overview-zh-CN.webp130 B
  • assets/screenshots/overview.webp130 B
  • assets/screenshots/thumbs/agent-detail-zh-CN.webp129 B
  • assets/screenshots/thumbs/agent-detail.webp129 B
  • assets/screenshots/thumbs/catalog-zh-CN.webp130 B
  • assets/screenshots/thumbs/catalog.webp130 B
  • assets/screenshots/thumbs/overview-zh-CN.webp129 B
  • assets/screenshots/thumbs/overview.webp129 B
  • busabase.json813 B
  • content/_folder.json105 B
  • content/agents/base.json1.9 KB
  • content/agents/records.ndjson1.5 KB
  • content/kelly-agent-builder-app/_node.json140 B
  • content/kelly-agent-builder-app/.busabaseignore194 B
  • content/kelly-agent-builder-app/app/accent-theme.css8.0 KB
  • content/kelly-agent-builder-app/app/accent-theme.js7.3 KB
  • content/kelly-agent-builder-app/app/app.js24.8 KB
  • content/kelly-agent-builder-app/app/base-ui.css13.0 KB
  • content/kelly-agent-builder-app/app/i18n/messages.js4.7 KB
  • content/kelly-agent-builder-app/app/index.html4.3 KB
  • content/kelly-agent-builder-app/app/js/agent-model.js5.5 KB
  • content/kelly-agent-builder-app/app/js/busabase-client.js226 B
  • content/kelly-agent-builder-app/app/js/config.js2.5 KB
  • content/kelly-agent-builder-app/app/js/connect-gate.js1.7 KB
  • content/kelly-agent-builder-app/app/js/providers/busabase-provider.js10.9 KB
  • content/kelly-agent-builder-app/app/js/providers/demo-provider.js4.8 KB
  • content/kelly-agent-builder-app/app/js/providers/index.js270 B
  • content/kelly-agent-builder-app/app/js/runtime.js3.4 KB
  • content/kelly-agent-builder-app/app/js/tool-catalog.js1.3 KB
  • content/kelly-agent-builder-app/app/styles.css16.8 KB
  • content/kelly-agent-builder-app/app/styles/editorial.css30.7 KB
  • content/kelly-agent-builder-app/app/vendor/busabase-airapp-gate.css6.1 KB
  • content/kelly-agent-builder-app/app/vendor/busabase-airapp-gate.js11.2 KB
  • content/kelly-agent-builder-app/app/vendor/busabase-airapp.js556.0 KB
  • content/kelly-agent-builder-app/app/vendor/busabase-sdk.js849.5 KB
  • content/kelly-agent-builder-app/package.json1.1 KB
  • content/kelly-agent-builder-app/pnpm-lock.yaml5.2 KB
  • content/kelly-agent-builder-app/pnpm-workspace.yaml50 B
  • content/kelly-agent-builder-app/scripts/check-editorial.mjs6.1 KB
  • content/kelly-agent-builder-app/scripts/check.mjs4.2 KB
  • content/kelly-agent-builder-app/server.js3.1 KB
  • content/kelly-agent-builder-app/test/agent-model.test.mjs3.1 KB
  • content/settings/base.json900 B
  • package.json292 B
  • pnpm-lock.yaml4.3 KB
  • pnpm-workspace.yaml50 B
  • README.md2.9 KB
  • references/agent-config-schema.md3.7 KB
  • scripts/publish_airapp.mjs3.1 KB
  • scripts/setup.mjs6.4 KB
  • scripts/sync-content.mjs2.5 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。

Agent Builder & Governance Console

Overview

Kelly Agent Builder is a Busabase Cloud App-in-Skill. Its canonical product surface is the AirApp in Busabase, not a separate local-data product. The same Hono source supports an explicitly requested local preview with OAuth connection bootstrap. It is a platform team's governance console for a catalog of mock agent configs. It never provisions or calls a real agent — every action reads or writes a Busabase Base record. This is a generic, brand-free tool: teams define a name, a trigger/intent description, a set of allowed tools (from a fixed catalog), an approval flag, and a monthly call quota, and this console tracks status and usage and gates the risky transition (draft → live) behind required-field validation.

Default behavior is AirApp-first. Unless the user explicitly asks only for explanation, update Busabase directly and give the user the clickable AirApp URL. Start localhost only when local preview/debugging is explicitly requested; it uses the same Busabase resources and never offers another data provider.

Mandatory Dependencies

  1. Read and follow $kelly-app-skill-creator for product behavior, visual quality, responsive layout, and the complete canonical content/kelly-agent-builder-app/ artifact.
  2. Read and follow $busabase for connection, target Space, node discovery, ChangeRequests, review, and merge behavior.
  3. Read and follow $busabase-app-creator for resource modeling, AirApp runtime limits, security, validation, and deployment.

If a dependency is unavailable, preserve this skill's local artifact and product contracts, stop before the unavailable Busabase operation, and report the exact missing dependency. Do not invent a second data backend.

App UI Screenshots

<table> <tr> <td width="50%"><img src="assets/screenshots/overview.webp" alt="Agent Builder overview"></td> <td width="50%"><img src="assets/screenshots/catalog.webp" alt="Agent Builder catalog"></td> </tr> <tr> <td><strong>Overview</strong><br>Governance summary: live agent count, aggregate quota usage, and a list of agents that need attention with reasons.</td> <td><strong>Catalog</strong><br>Sortable, searchable table of every agent config with status badges, owning team, and quota usage.</td> </tr> <tr> <td width="50%"><img src="assets/screenshots/agent-detail.webp" alt="Agent Builder agent detail"></td> <td width="50%"></td> </tr> <tr> <td><strong>Agent detail / edit</strong><br>Tool checklist, quota input, approval toggle, owning team field, trigger/intent textarea, status, and lifecycle actions (activate / pause / archive).</td> <td></td> </tr> </table>

Boundary

  • This is a mock governance console. It never provisions, deploys, or calls any real agent, model, or external tool. The "allowed tools" checklist is a fixed local catalog (content/kelly-agent-builder-app/app/js/tool-catalog.js) used only for governance bookkeeping — selecting a tool here does not grant or invoke it anywhere.
  • The AirApp reads and writes Busabase records only; it must not call any other remote system.
  • No brand-specific integration exists or is implied.

Busabase Resources

Two Bases under one application Folder (kelly-agent-builder), declared in content/kelly-agent-builder-app/app/js/config.js and the generated template sidecars under content/:

  • agents: agent-id, name, trigger-description, allowed-tools (JSON array), approval-required, monthly-quota, calls-this-month, owning-team, status, created-at, updated-at.
  • settings: one row per kind — kelly-agent-builder-onboarding (presence marks setup complete; this skill has no external accounts or secrets to configure) and kelly-agent-builder-lock.

Resources provision lazily through an idempotent Busabase ChangeRequest the first time the app runs in a Space; see references/agent-config-schema.md for exact field shapes and governance rules.

Authentication

Busabase authentication is ambient inside the deployed AirApp, which must not show OAuth, API-key, Base URL, provider, hosting, or Space controls. Standalone loopback preview uses browser OAuth without exposing tokens; after OAuth it auto-selects a single/open-source Space or requires a native selector when several Spaces are accessible.

Demo Mode

  • ?demo=1 opens a deterministic, fully offline mock catalog (8 agent configs spanning draft/live/paused/archived, one over-quota, one missing an owning team) for documentation and screenshots.
  • lang=en or lang=zh forces UI chrome language for screenshots.
  • Demo mode never reads or writes Busabase.

Governance Rules

Read references/agent-config-schema.md before editing the app or its domain logic. In short:

  • Draft → live is only allowed when name, trigger_description, at least one allowed_tools entry, non-empty owning_team, and monthly_quota > 0 are all present. This is enforced in content/kelly-agent-builder-app/app/js/providers/busabase-provider.js#activateAgent — the browser form disables the button too, but the provider is the source of truth.
  • Archive is allowed from any status. Archived agents become read-only.
  • Pause is only allowed from live.
  • Needs attention = a draft with missing required fields, OR an agent (any status) with no owning team, OR a quota-reached live agent (calls_this_month >= monthly_quota — reached, not strictly exceeded), OR approval_required: true with no owning team assigned.
  • Update validation: an edit that would leave an already-live agent missing any required field (e.g. clearing owning_team or allowed_tools) is rejected, the same gate activate uses. Draft agents remain freely editable.

Local App

  • content/kelly-agent-builder-app/app/index.html + content/kelly-agent-builder-app/app/app.js + content/kelly-agent-builder-app/app/styles.css + content/kelly-agent-builder-app/app/i18n/messages.js: zero-build vanilla frontend with hash routing (#/overview, #/catalog, #/agent/:id, #/agent/new, #/settings).
  • content/kelly-agent-builder-app/app/js/agent-model.js: pure governance rules (missingRequiredFields, isQuotaReached, deriveAgent, summarize, lifecycle transitions) shared by the busabase and demo providers.
  • content/kelly-agent-builder-app/app/js/providers/: busabase-provider.js (reads/writes via busabase-sdk) and demo-provider.js (deterministic, read-only).
  • content/kelly-agent-builder-app/app/js/tool-catalog.js: the fixed tool catalog (web_search, code_exec, file_read, file_write, send_email, calendar, crm_lookup, db_query, slack_post, http_request).
  • content/kelly-agent-builder-app/server.js: thin Hono OAuth bootstrap + same-origin /api/v1 proxy to Busabase — no business logic.

Safety

  • Never provision or call a real agent, tool, or external system from this skill's app.
  • Keep owning_team values as free text; do not validate against a real directory service.

Completion Criteria

Finish only when:

  • the skill contains the complete canonical content/kelly-agent-builder-app/ project and pnpm --dir content/kelly-agent-builder-app dev remains supported;
  • all persistent config and agent data use busabase-sdk and the declared resource map — no local JSON, browser storage, or provider choice;
  • local setup offers Cloud/custom URL OAuth plus the explicit Demo path, while a deployed AirApp uses its ambient session;
  • Overview, Catalog, Agent detail, and Help & Settings render on desktop and phone widths;
  • pnpm --dir content/kelly-agent-builder-app run check and node --test pass.

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

Align all AI coding agents to a single source of truth. AGENTS.md for rules, .agents/skills/ for skills, symlinks for everything else. Use when setting up a repo for multiple agents, or when links are missing, broken, or out of date.

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。

mr-kelly/skills52026年10月2日 更新

Read, answer and post on the Buda community forum at community.buda.im through its /api/v1/community REST API. Lists categories and posts, reads a thread with its replies, searches, and publishes posts and replies behind an explicit confirmation. Use when the user invokes $buda-community or /buda-community, asks what is happening in the Buda community, wants an unanswered question triaged or answered, or wants something posted or announced there.

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。

mr-kelly/skills52026年10月2日 更新

Read, answer and post on the Busabase community forum at community.busabase.com through its /api/v1/community REST API. Lists categories and posts, reads a thread with its replies, searches, and publishes posts and replies behind an explicit confirmation. Use when the user invokes $busabase-community or /busabase-community, asks what is happening in the Busabase community, wants an unanswered question triaged or answered, or wants something posted or announced there.

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。

mr-kelly/skills52026年10月2日 更新

kelly-ads

無料

Ad-campaign command desk (投放指挥台, Busabase App-in-Skill) for a cross-border e-commerce seller running Amazon Ads, Meta (FB/IG), TikTok Ads, and Google Ads. Use when the user invokes $kelly-ads or /kelly-ads, or asks about 投放, ad campaigns, ad spend, ACOS, ROAS, Amazon Ads, TikTok ads, Meta ads, Google Ads, search terms, negative keywords, bid adjustment, budget pacing, CPC spikes, rejected ads, ad anomaly detection, 广告优化, or reviewing agent-proposed adjustment cards (negative keyword, bid down/up, pause target, budget shift, creative refresh).

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。

mr-kelly/skills52026年10月2日 更新

Review board (Busabase App-in-Skill) that runs a fixed suite of mock test cases against a baseline vs candidate agent version and surfaces rubric-scored regressions before a release. Use when the user invokes $kelly-agent-eval or /kelly-agent-eval, wants to review agent-version regressions, compare baseline vs candidate quality, triage a release, or record a release approve/block decision. Deterministic mock rubric scores only — not a real LLM-judge call, and it never deploys anything.

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。

mr-kelly/skills52026年10月2日 更新

Busabase App-in-Skill dashboard visualizing a generated MOCK fleet of LLM agents running behind a shared AI gateway for a generic organization. Use when the user invokes $kelly-agent-observability or /kelly-agent-observability, wants to review agent health, call volume, latency, error rates, cost, or trace-level chain breaks across an agent fleet. This is a demo/reference dashboard over a generated mock fleet snapshot seeded into Busabase by a trusted script — never a live monitoring integration, never a real gateway or real agents.

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。

mr-kelly/skills52026年10月2日 更新

mr-kelly のスキルをすべて見る

このスキルの問題を報告する