How do I install the Co Consultant Acn Deep Research skill?

Run npx skillmds@latest add nahisaho/co-consultant-acn-deep-research in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Co Consultant Acn Deep Research safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Co Consultant Acn Deep Research?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Co Consultant Acn Deep Research free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Co Consultant Acn Deep Research?

nahisaho (@nahisaho) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Co Consultant Acn Deep Research by nahisaho

Name: Co Consultant Acn Deep Research
Author: nahisaho

Co Consultant Acn Deep Research

Iterative deep research with Think→Action→Report cycle for Accenture-style engagements. Web search via web_search tool, /research command, and MCP integration. Bilingual search, source quality assessment, and cross-validation. Use when COLLECTING information, verifying market data, conducting competitive analysis, or gathering evidence for ACN framework application.

nahisaho Updated Sep 17, 2026

File contents

Deep Research (ACN)

Hypothesis-driven iterative research for Accenture-style consulting.

Use This Skill When

Collecting market, competitive, or technology intelligence.
Verifying business data from multiple sources.
Gathering evidence before applying ACN frameworks.

Research Tools (Priority Order)

1. `web_search` tool (Built-in — Always Available)

Targeted web queries. Always run bilingual (JP + EN):

web_search("Japan digital transformation market size 2025")
web_search("Accenture ACN framework case study")

Best for: focused data points, market statistics, recent news, company data.

2. `/research` command (CLI — Interactive Sessions)

Comprehensive multi-source investigation:

/research What is the competitive landscape for [target market]?

Best for: complex research questions needing automated sub-question decomposition.

3. `deep-research` MCP (When Configured)

Full pipeline: elaborate → subquestions → search → analyze → report.

Workflow

Think Phase

Analyze knowledge gaps against the objective.
Generate hypotheses.
Choose tool: web_search for targeted, /research for comprehensive, MCP for pipeline.

Action Phase

Action	Tool	When
SEARCH	`web_search` (bilingual JP+EN)	Need specific data points
RESEARCH	`/research` command	Need comprehensive multi-source investigation
VISIT	`web_fetch`	Need full page content from URL
VERIFY	`web_search` (2nd query)	Cross-validate key numbers/facts
COMPLETE	—	All completion criteria met

Bilingual: Always search in both Japanese and English.
Academic priority: Prefer .ac.jp, .edu, .gov domains.
Cross-validation: Verify key data with 2+ sources.

Report Phase

Update findings after each round.
Tag each fact with source URL and trust level.
Mark single-source findings with ⚠️.

Deliverables

results/research-notes.md: findings with source attribution.
results/source-registry.md: all sources with trust ratings.

Quality Gates

3+ trusted sources collected.
Weighted trust average ≥ 60%.
Key hypotheses verified or refuted.
Single-source claims marked with ⚠️.

If any gate fails: identify the issue, fix, and re-validate.

Gotchas

web_search は常に利用可能。MCP が設定されていなくても Web 検索は実行できる
検索は日英両方で行うこと。英語のみだと日本市場固有のデータを見落とす
/research コマンドは CLI インタラクティブセッションでのみ使用可能。自動化では web_search を使う
Accentureのレポート・出版物は信頼度「高」として扱うが、自社バイアスに注意すること
収集データは各ラウンド後にファイルに保存。コンパクションで消失する
Phase間の引き継ぎは必ず results/research-notes.md に保存してからハンドオフすること

Validation Loop

リサーチラウンドを実行
チェック: 3+ソース、交差検証済み、⚠️タグ付き
不合格なら追加ラウンド
全基準を満たしてからCOMPLETE

---name: co-consultant-acn-deep-researchdescription: Iterative deep research with Think→Action→Report cycle for Accenture-style engagements. Web search via web_search tool, /research command, and MCP integration. Bilingual search, source quality assessment, and cross-validation. Use when COLLECTING information, verifying market data, conducting competitive analysis, or gathering evidence for ACN framework application.---# Deep Research (ACN)Hypothesis-driven iterative research for Accenture-style consulting.## Use This Skill When- Collecting market, competitive, or technology intelligence.- Verifying business data from multiple sources.- Gathering evidence before applying ACN frameworks.## Research Tools (Priority Order)### 1. `web_search` tool (Built-in — Always Available)Targeted web queries. Always run bilingual (JP + EN):```web_search("Japan digital transformation market size 2025")web_search("Accenture ACN framework case study")```Best for: focused data points, market statistics, recent news, company data.### 2. `/research` command (CLI — Interactive Sessions)Comprehensive multi-source investigation:```/research What is the competitive landscape for [target market]?```Best for: complex research questions needing automated sub-question decomposition.### 3. `deep-research` MCP (When Configured)Full pipeline: elaborate → subquestions → search → analyze → report.## Workflow### Think Phase- Analyze knowledge gaps against the objective.- Generate hypotheses.- Choose tool: `web_search` for targeted, `/research` for comprehensive, MCP for pipeline.### Action Phase| Action | Tool | When ||--------|------|------|| SEARCH | `web_search` (bilingual JP+EN) | Need specific data points || RESEARCH | `/research` command | Need comprehensive multi-source investigation || VISIT | `web_fetch` | Need full page content from URL || VERIFY | `web_search` (2nd query) | Cross-validate key numbers/facts || COMPLETE | — | All completion criteria met |- **Bilingual**: Always search in both Japanese and English.- **Academic priority**: Prefer `.ac.jp`, `.edu`, `.gov` domains.- **Cross-validation**: Verify key data with 2+ sources.### Report Phase- Update findings after each round.- Tag each fact with source URL and trust level.- Mark single-source findings with ⚠️.## Deliverables- `results/research-notes.md`: findings with source attribution.- `results/source-registry.md`: all sources with trust ratings.## Quality Gates- [ ] 3+ trusted sources collected.- [ ] Weighted trust average ≥ 60%.- [ ] Key hypotheses verified or refuted.- [ ] Single-source claims marked with ⚠️.If any gate fails: identify the issue, fix, and re-validate.## Gotchas- `web_search` は常に利用可能。MCP が設定されていなくても Web 検索は実行できる- 検索は日英両方で行うこと。英語のみだと日本市場固有のデータを見落とす- `/research` コマンドは CLI インタラクティブセッションでのみ使用可能。自動化では `web_search` を使う- Accentureのレポート・出版物は信頼度「高」として扱うが、自社バイアスに注意すること- 収集データは各ラウンド後にファイルに保存。コンパクションで消失する- Phase間の引き継ぎは必ず `results/research-notes.md` に保存してからハンドオフすること## Validation Loop1. リサーチラウンドを実行2. チェック: 3+ソース、交差検証済み、⚠️タグ付き3. 不合格なら追加ラウンド4. 全基準を満たしてからCOMPLETE

Frequently asked questions