Borrowing it
Nothing to install: this file belongs to geekfujiwara/CodeAppsDevelopmentStandard. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/geekfujiwara/CodeAppsDevelopmentStandard/main/.github/skills/copilot-studio-v2/SKILL.mdgit clone --depth 1 https://github.com/geekfujiwara/CodeAppsDevelopmentStandardWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/geekfujiwara/codeappsdevelopmentstandard/copilot-studio-v2)<a href="https://agentmods.dev/skills/geekfujiwara/codeappsdevelopmentstandard/copilot-studio-v2"><img src="https://agentmods.dev/badge/skills/geekfujiwara/codeappsdevelopmentstandard/copilot-studio-v2/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/geekfujiwara/codeappsdevelopmentstandard/copilot-studio-v2"><img src="https://agentmods.dev/badge/skills/geekfujiwara/codeappsdevelopmentstandard/copilot-studio-v2.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Privilege Escalation · line 78 Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
- high Privilege Escalation · line 303 Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00113 | $0.07842 |
| Opus 5.5 | $0.00045 | $0.03137 |
| Sonnet 5.5 | $0.00023 | $0.01568 |
| Haiku 4.5 | $0.00011 | $0.00784 |
Grade A, and why
copilot-studio-v2 scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 14d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 403 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Copilot Studio v2（新アーキテクチャ）エージェント構築スキル
Copilot Studio の 「全く新しいアーキテクチャ」（cliagent テンプレート） エージェントを、
Dataverse Web API と、必要時はログイン済みブラウザの observed/private API で構築する。
標準方針: 新規の Copilot Studio 開発は v2 を第一候補にする。Code Apps では Web app チャネルの iframe を表示でき、Agent flows では既存の発行済み v2 を Agent ノードから利用できる。 iframe 表示、ブラウザー SDK によるプログラム的な会話制御、Workflow のサーバー側呼び出しは 異なる連携方式であり、可否・認証・受入を分けて判定する。
v1（旧）スキルとの最大の違い
| 観点 | v1（copilot-studio スキル / 旧アーキ） |
v2（本スキル / 新アーキ cliagent） |
|---|---|---|
| Bot 作成 | ❌ API 不可。Copilot Studio UI で手動作成必須 | POST /bots を第一候補にし、pac copilot list に現れない場合だけ UI 作成へ fallback |
| 設定の保存先 | GPT コンポーネント（componenttype=15）+ PVA ダブル改行 YAML | bots.configuration の BotConfiguration JSON にインライン |
| recognizer | （クラシック PVA） | CLICopilotRecognizer |
| モデル指定 | GPT data の aISettings.model.modelNameHint |
Auto が標準。model を省略してプラットフォームに任せ、固定モデルのみ series 指定 |
| Instructions | GPT data YAML（ダブル改行フォーマット注意） | agentSettings.instructions.segments[].value（プレーン文字列） |
| メモリ | （個別設定） | agentSettings.enableMemory: true |
| スキル | （ナレッジ/トピック） | フラット Python スキルバンドル（type=9 + type=14 子ファイル） |
| 自動化適性 | △ UI 介在が必要 | ◎ 基本構築 + 承認済み初回ツール投入。OAuth/consent は browser session |
このスキルを選ぶ理由: Bot 作成からスキル添付まで 人手の UI 操作ゼロ で構築できる。 CI/再現構築・量産・プログラム的な改変に向く。
いつ v2 を使うか（architecture スキルでの分岐）
architecture スキルの Copilot Studio 選定では、新規開発は v2 を標準の第一候補にする。
v1 は既存 v1 資産の継続、または v2 で満たせない直接 SDK 要件がある場合に限って選ぶ。
| 利用方式 | 標準選択 | 境界 |
|---|---|---|
| Teams / Copilot Studio で対話 | v2 | 公開先ごとに認証と本人権限を受入確認する |
| Code Apps に標準チャット画面を表示 | v2 Web app iframe | iframe 内 UI。親アプリからの任意メッセージ注入、応答イベント取得、SSO/OBO を自動的には保証しない |
| Dataverse レコード作成・更新から自動実行 | Agent flows + v2 Agent ノード | Workflow の接続所有者、ACP/DLP、実行完了、業務出力を個別に検証する |
| Code Apps で要求・結果を業務 UI と統合 | Dataverse + Agent flows + v2 | 非同期の要求/結果方式。相関、Claim、行所有、再実行防止が必要 |
| 親アプリで会話をプログラム制御 | 要件確認後に v1 | ExecuteCopilotAsyncV2、WebChat SDK、トークンストリーミング等が必須の場合 |
Web app iframe の表示可否と、ExecuteCopilotAsyncV2 / WebChat SDK による直接呼び出し可否を混同しない。
v2 iframe は実際の公開 URL、Code Apps の frame-src、サインイン、一般利用者の本人認可を公開ホストで確認する。
Agent flows の既存 Agent ノードは製品 UI のサポート経路を使い、直接 API Hub 呼び出しを標準実装にしない。
構築・検証は agent-flows に従う。
What ships with it
50 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/.env.example 5.4 KB
- references/app-details.md 6.8 KB
- references/connector-tools.md 4.3 KB
- references/data-agent-skill-authoring.md 9.5 KB
- references/draft-only-worker-agent.md 2.8 KB
- references/flat-python-skill.md 5.5 KB
- references/icon-and-publish.md 4.0 KB
- references/mcp-servers.md 10 KB
- references/model-series.md 3.3 KB
- references/new-architecture.md 4.5 KB
- references/skill-bundle-structure.md 4.0 KB
- references/troubleshooting.md 9.1 KB
- scripts/agent_env.py 3.2 KB runs code
- scripts/analyze_agent.py 3.2 KB runs code
- scripts/attach_skill.py 6.2 KB runs code
- scripts/connector_tool_verify.py 3.2 KB runs code
- scripts/create_agent.py 9.8 KB runs code
- scripts/create_initial_tools_manifest.py 2.7 KB runs code
- scripts/deploy_agent.py 7.2 KB runs code
- scripts/mcp_tool_browser_runner.mjs 17 KB runs code
- scripts/mcp_tool_plan.py 7.1 KB runs code
- scripts/model_policy.py 1.5 KB runs code
- scripts/publish_agent.py 4.3 KB runs code
- scripts/set_app_details.py 18 KB runs code
- scripts/set_icon.py 6.4 KB runs code
- scripts/set_instructions.py 3.7 KB runs code
- scripts/set_model.py 5.2 KB runs code
- scripts/set_prompts.py 6.4 KB runs code
- scripts/update_agent.py 6.9 KB runs code
- scripts/verify_agent.py 2.8 KB runs code
- scripts/verify_config.py 1.8 KB runs code
- templates/agm-qa-author/agent.env 1.4 KB
- templates/agm-qa-author/instructions.md 2.8 KB
- templates/agm-qa-author/prompts.json 707 B
- templates/agm-qa-author/README.md 6.0 KB
- templates/agm-qa-author/scaffold.json 3.2 KB
- templates/agm-qa-author/skills/agm-qa-authoring/SKILL.md 10.0 KB
- templates/agm-qa-author/skills/agm-qa-review/SKILL.md 5.1 KB
- templates/agm-qa-author/skills/agm-rehearsal-script/SKILL.md 8.9 KB
- tests/test_agent_env.py 3.4 KB runs code
- tests/test_agm_qa_author_template.py 4.1 KB runs code
- tests/test_connector_tool_verify.py 2.8 KB runs code
- tests/test_create_initial_tools_manifest.py 2.7 KB runs code
- tests/test_deploy_agent.py 2.3 KB runs code
- tests/test_mcp_tool_browser_runner.mjs 8.3 KB runs code
- tests/test_mcp_tool_plan.py 4.8 KB runs code
- tests/test_model_policy.py 8.0 KB runs code
- tests/test_publish_agent.py 1.4 KB runs code
- tests/test_set_icon.py 2.4 KB runs code
- tests/test_update_agent.py 686 B runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 14d ago Changed · +11 lines b2c2d0c3aff9
- 1mo ago Changed 81e0ca2eec57
- 1mo ago Changed · +17 lines 605ae5f8a692
- 1mo ago Changed · +27 lines · -11 tokens per session 90be4e2f81f0
- 1mo ago Changed · +17 lines f67f0be352b6
- 1mo ago Changed · +4 lines 756d6c839ded
- 1mo ago Changed · +5 lines 75f093a80a8e
- 2mo ago Changed · +13 lines 5e5a91cd7ff1
- 2mo ago First seen · 309 lines · 124 tokens per session scan A ea44f519546b
copilot-studio-v2 is a skill published in the GitHub repository geekfujiwara/CodeAppsDevelopmentStandard (78 stars, last pushed yesterday), licensed MIT. It adds 113 tokens to every session and 7,842 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
copilot-studio-v2 compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| copilot-studio-v2geekfujiwara/CodeAppsDevelopmentStandard | Skill | 78 | yesterday | MIT | A | 113 | Claude Code, Codex |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 103 | Claude Code |