SuperAPI AI Directory
Partner with us中文AI newsFavorites
Skillsenterprise-ai-scenario-map

enterprise-ai-scenario-map

General & other

View repository

Updated 10-16 00:29

About this skill

Enterprise AI Scenario Map Generator

Task Objective

  • This Skill is used to: generate an AI application scenario map report for an enterprise, helping decision-makers quickly identify applicable AI scenarios and priorities
  • Core value: let business owners quickly know which scenarios in their company can use AI and which should be launched first
  • Capabilities include:
    1. In-depth enterprise research (company information, main business, products, industry tags)
    2. Business characteristics and process analysis
    3. Search for AI best practice cases in the same industry
    4. AI application scenario map generation (full inventory of 30+ scenarios, including priority recommendations)
    5. Implementation roadmap planning (phased rollout plan)
  • Trigger conditions: the user needs to formulate an AI implementation plan for a certain enterprise, or wants to understand "how my company can use AI"

Prerequisites

  • Dependency notes:
    • The agent needs to have web-search(web search) tool capability
    • Python 3.8+ (used to run the research framework generation script)
    • No third-party Python package dependencies

Operation Steps

Core workflow (strictly execute in order)

Important: search must be completed before entering the analysis and report generation stage

Phase 1: In-depth research (data collection)

This phase must be completed first before entering the analysis phase

Step 1.1: Generate research framework

Run the script to generate the research framework and search question list:

python scripts/deep_research_wrapper.py --company-name "<公司名称>" --country "<国家>"

The script will output:

  • Enterprise research framework (basic structure)
  • Enterprise information search question list (8 core questions)
  • Industry information search question list (pain points + cases)

Step 1.2: In-depth enterprise information research

Use the web-search tool to search one by one the "Enterprise Information Search Question List" output in step 1.1, and collect the following information:

  • Basic company information (founding time, scale, main business)
  • Business model (B2B/B2C/platform-based/hybrid)
  • Core business links
  • Key business processes
  • Customer types
  • Business characteristics
  • Team size (if public information is available)

Step 1.3: Industry pain point and case collection

Use the web-search tool to search according to the "Industry Information Search Question List" output in step 1.1 (must be in order):

Search A: Common industry pain points Search keywords:<行业> + 痛点 + 挑战 + 2024 2025 Collect content:

  • Common pain points faced by the industry
  • Typical manifestations and impacts
  • Quantitative data (if available)

Search B: Industry AI application cases Search keywords:<行业> + AI应用 + 智能化 + 案例 Search keywords:<具体业务> + AI + LLM + 实践 Collect content:

  • At least 3 benchmark enterprise cases
  • Each case includes: enterprise background, application scenario, technical solution, implementation effect, key success factors

Phase 1 completion marker: all searches are completed, and the information has been organized into a structured document

Phase 2: Analysis and Diagnosis (start only after Phase 1 is completed)

Step 2.1: Business Characteristics Analysis Refer to references/business-analysis-framework.md, analyze:

  • Overview of business characteristics (service model, professional requirements, business collaboration, knowledge dependency, customer characteristics)
  • Internal process characteristics (process characteristics 1-4 and AI empowerment directions)

Step 2.2: Core Pain Point Diagnosis Based on the pain point information collected in Phase 1, analyze:

  • Common industry pain points (4)
  • Enterprise operational pain points (five dimensions: efficiency, quality, risk, cost, knowledge)

Step 2.3: Summary of Benchmarking Insights Based on the collected industry cases, summarize:

  • 4 key insights
  • Explanation and recommended actions for each insight

Phase 3: Scenario Map Generation (start only after Phase 2 is completed)

Step 3.1: Business Process Breakdown Based on the enterprise's main business, break down the core business processes, format example:

项目立项 → 预算编制&造价 → 招标代理&评标 → 合同管理&变更 → 工程结算&审计

Step 3.2: AI Scenario Full Inventory Generation (30+ scenarios) Refer to references/typical-ai-scenarios.md, generate a full scenario inventory, including the following columns: | No. | Business Segment | AI Scenario Name | Function Description | Implementation Prerequisites | Expected Benefits | Priority |

Scenario Generation Requirements：

  • The total number of scenarios must reach more than 30
  • Classified by business segment (at least 5 segments)
  • Each scenario must include: function description, implementation prerequisites, expected benefits, priority (🟢/🟡/🔵)

Priority Definitions(Refer to references/scenario-priority-framework.md）：

  • 🟢 Quick Start (0-3 months): mature technology, fast implementation, quick results
  • 🟡 Mid-term Build (3-12 months): requires a certain foundation, high value
  • 🔵 Long-term Evolution (over 1 year): requires deep accumulation, high strategic value

Step 3.3: Priority Matrix Analysis Classify scenarios by "business value" and "implementation feasibility":

  • Priority Implementation Zone (high value + high feasibility)
  • Strategic Investment Zone (high value + medium feasibility)
  • Quick Win Zone (medium value + high feasibility)
  • Other zones...

Step 3.4: In-depth Interpretation of Key Scenarios Select the 3 highest-priority scenarios and explain in detail:

  • Scenario Name
  • Scenario Positioning
  • Business Pain Points
  • AI Solution
  • Expected Benefits
  • Implementation Prerequisites
  • Implementation Cycle
  • Success Cases (if any)

Phase 4: Report Generation (strictly follow the V2.1 template format)

Step 4.1: Generate the Complete Report Generate the report according to the following structure (strictly follow the V2.1 template):

Cover

  • Client Name
  • Report Date
  • Prepared by: Yuanzi (Shenzhen) Artificial Intelligence Co., Ltd. (fixed, not replaceable)

Part 1: Executive Summary (1 page)

  • About the Client Company
  • Strategic Value of AI Empowerment (C-level perspective)
    • Strategic Value Proposition (three major strategic values: competitive advantage upgrade, significant financial returns, sustainable growth)
    • Strategic Value Quantification Table (comparison of 5 strategic dimensions)
    • Strategic opportunity windows (time window, late-mover advantage, differentiation opportunities)
  • Three core challenges identified (in table form)
  • Core value proposition of AI empowerment
  • Three scenarios recommended for priority launch (in table form, including 🥇🥈🥉)

Part 2: Enterprise Profile and Business Diagnosis (3-4 pages) 2.1 Enterprise Snapshot (table) 2.2 Business Value Chain Overview (table, including key activities and support systems) 2.3 Business Characteristics Analysis (business characteristics overview table, internal process characteristics table) 2.4 Core Pain Point Diagnosis (industry common pain points table, enterprise operational pain points table)

Part 3: Industry AI Practice Scan (2 pages) 3.1 What Industry Benchmarks Have Done (comparison table of 3 cases) 3.2 Benchmarking Insights: What Can Be Learned (table of 4 insights)

Part 4: AI Scenario Map (5-6 pages) 4.1 Business Process and AI Scenario Overview (table, including scenario count and priority distribution) 4.2 Full AI Scenario List (table of 30+ scenarios) 4.3 Scenario Statistics Summary (statistics dimension table) 4.4 Priority Matrix (3x3 matrix table) 4.5 In-Depth Interpretation of Key Scenarios (3 scenario cards)

Part 5: Implementation Path Recommendations (2 pages) 5.1 Phased Implementation Plan (comparison table of Phase 1, Phase 2, and Phase 3) 5.2 Key Success Factors (table of 5 factors)

Part 6: How We Can Help You (1 page) 6.1 Service Capability Matrix (table of 4 service types) 6.2 Next-Step Action Recommendations (table of 4 steps) 6.3 Contact Us(use fixed information)

  • Prepared by: Yuanzi (Shenzhen) Artificial Intelligence Co., Ltd.
  • Contact: Neil
  • WeChat ID: 838000702
  • WeChat Official Account: 源子AI Metainflow

Appendix: Template Usage Guide

  • A. How to Use This Template
  • B. Quality Checklist
  • C. Handling Common Issues
  • D. Scenario Expansion Reference

Important: Company Information Is Fixed

  • The "Prepared by" on the cover uses fixed information and cannot be replaced
  • The "Contact Us" in Part 6 uses fixed contact information
  • Reference:references/company-info-config.md

Step 4.2: Quality Check After generating the report, verify against the following checklist items:

  • Is the executive summary within 1 page?
  • Is the business characteristics analysis sufficient?
  • Are the pain point descriptions specific and quantifiable?
  • Is there a clear correspondence between scenarios and pain points?
  • Does the scenario list reach 30+ items?
  • Is the priority ranking persuasive?
  • Are the industry cases relevant to the client's business?
  • Is the implementation path clear and executable?
  • Does it include all 6 complete Parts?

Optional Branches

Branch A: Quick Scan Mode Applicable scenario: The user wants to quickly understand AI application potential Execution strategy:

  • Phase 1: Conduct only basic enterprise information research, skip in-depth business information collection
  • Phase 2: Simplify business analysis and pain point diagnosis
  • Phase 3: Generate only 15-20 core scenarios (only 🟢 priority)
  • Phase 4: Generate a simplified report (only Part 1, Part 2, and Part 4 summary)

Branch B: In-Depth Planning Mode Applicable scenario: The user needs to formulate a detailed implementation plan Execution strategy:

  • Phase 1: Complete all research and information collection, expand to 10+ industry cases
  • Phase 2: Add competitor analysis and customer interview simulation
  • Phase 3: Generate a full list of 50+ scenarios, including four priority levels: P0/P1/P2/P3
  • Phase 4: Generate the full report + detailed implementation plan + ROI calculation

Branch C: Specific Domain Focus Mode Applicable scenario: The user focuses on a specific business domain (such as customer service, R&D, marketing) Execution strategy:

  • Phase 1: Enterprise research, with emphasis on collecting information in that domain
  • Phase 2: Analysis focuses on that domain
  • Phase 3: Scenarios in that domain account for more than 60% of the full scenario list
  • Phase 4: Highlight that domain in the report; all in-depth interpretations of key scenarios in Part 4 are selected from that domain

Resource Index

Required Scripts

  • scripts/deep_research_wrapper.py
    • Purpose: Generate the enterprise research framework and search question list
    • Parameters:--company-name(required),--country(default "China"),--format(markdown/json, default markdown)
    • Dependencies: Python 3.8+, no third-party package dependencies

Domain References

  • references/company-info-config.md
    • When to read: Always (fixed company information configuration)
    • Content: Yuanzi AI company information configuration (name, contact person, WeChat), automatically used in reports
  • references/report-template-v2.1.md
    • When to read: Phase 4, Step 4.1
    • Content: Complete V2.1 report template, including detailed formatting requirements and examples for 6 Parts
  • references/business-analysis-framework.md
    • When to read: Phase 2, Step 2.1
    • Content: Business characteristics analysis framework, process characteristics analysis dimensions
  • references/industry-case-template.md
    • When to read: Phase 1, Step 1.3
    • Content: Industry case output format template, field completion instructions
  • references/scenario-priority-framework.md
    • When to read: Phase 3, Step 3.2
    • Content: Priority evaluation criteria (P0/P1/P2), evaluation process, evaluation tool table
  • references/typical-ai-scenarios.md
    • When to read: Phase 3, Step 3.2
    • Content: Reference library of typical AI scenarios for 8 major industries (construction, e-commerce, finance, manufacturing, healthcare, education, legal, logistics)

Output Assets

  • No fixed template; the report is dynamically generated by the agent based on the V2.1 template and user cases

Notes

Execution Order Requirements (Mandatory)

  1. All searches in Phase 1 (in-depth research) must be completed first, before entering Phase 2 analysis
  2. Phase 2 (analysis and diagnosis) must be completed first, before entering Phase 3 scenario generation
  3. Phase 3 (scenario map) must be completed first, before entering Phase 4 report generation
  4. It is strictly prohibited to skip the search phase and directly generate the report

Tool Dependency Requirements

  1. web-search: The agent must have web search capabilities for enterprise information research and industry case collection
  2. The script only generates the research framework; the actual research work is completed by the agent through web-search

Content Quality Requirements

  1. Number of scenarios: The full list of AI scenarios must reach more than 30
  2. Industry cases: At least 3 benchmark cases related to the client's business must be collected
  3. Pain point quantification: All pain point descriptions must include specific, quantifiable impacts
  4. Reasonable prioritization: Priority ranking must have sufficient basis (business value, implementation feasibility, expected benefits)
  5. Strict formatting: The report must be generated strictly according to the 6-Part structure of the V2.1 template

Data verification requirements

  1. Enterprise information must be cross-verified through at least two sources
  2. Industry cases must come from authoritative sources (official reports, industry reports, well-known media)
  3. Pain point and benefit data must have specific numbers or quantifiable descriptions
  4. Technical solution descriptions must be accurate and professional

Report generation principles

  1. Customer perspective: The entire report must be customer-centric, making the customer feel that "you understand my business"
  2. Value-oriented: Each scenario must clearly state the expected benefits and value proposition
  3. Actionability: The implementation path and recommendations must be clear and executable
  4. Differentiated guidance: Naturally guide toward consulting services, avoiding hard selling

Usage examples

Example 1: Complete report generation for a construction engineering company

Function description: Generate a complete AI scenario map report for "Shenzhen Hangjian Engineering Cost Consulting Co., Ltd."

Execution method：

  1. Phase 1: In-depth research

    • Execution script:python scripts/deep_research_wrapper.py --company-name "深圳市航建工程造价咨询有限公司" --country "中国"
    • Use web-search to search the question list output by the script one by one
    • Search "construction engineering industry pain points"
    • Search "engineering cost AI application cases", "bidding agency intelligence", etc.

  2. Phase 2: Analysis and diagnosis

    • Analyze business characteristics: whole-process investment control, high professionalism, multi-business collaboration
    • Diagnose core pain points: efficiency bottlenecks, difficulty in experience transfer, high compliance risks
    • Summarize benchmarking insights: data foundation, standardization entry point, human-machine collaboration, scenario-driven

  3. Phase 3: Scenario map

    • Business process breakdown: project initiation → budget preparation & cost estimation → bidding agency & bid evaluation → contract management & changes → project settlement & audit
    • Generate 32 AI scenarios (7 for budget preparation, 5 for bidding agency, 5 for contract management, 6 for project settlement, 7 for support systems)
    • Priority classification: 🟢 9 quick starts, 🟡 22 mid-term builds, 🔵 1 long-term evolution
    • Key scenarios: AI cost knowledge base, AI bid document intelligent review, AI policy and regulation analysis

  4. Phase 4: Report generation

    • Generate a complete V2.1 format report (6 Parts)
    • Part 1 Executive Summary: 3 core challenges, 3 priority launch scenarios (🥇AI cost knowledge base, 🥈AI bid document intelligent review, 🥉AI policy and regulation analysis)
    • Part 4 includes the full table of 32 scenarios and the priority matrix

Key outputs：

  • Complete report: about 15-20 pages
  • Scenario list: 32 AI application scenarios
  • Implementation roadmap: 3-phase rollout plan

Example 2: Rapid scan for an e-commerce company

Function description: Quickly generate an AI application scenario summary for an e-commerce platform

Execution method: Use rapid scan mode

  • Phase 1: Basic enterprise research (web-search)
  • Phase 2: Simplified analysis
  • Phase 3: Generate only 18 🟢 priority scenarios
  • Phase 4: Generate a simplified report (only Part 1, Part 2, and Part 4 summary)

Output: 5-8 page summary report

Example 3: In-depth planning for a manufacturing enterprise

Function description: Develop a detailed AI implementation plan for a manufacturing enterprise

Execution method: Use in-depth planning mode

  • Phase 1: Expand to 10+ industry cases (web-search)
  • Phase 2: Add competitor analysis
  • Phase 3: Generate 50+ scenarios
  • Phase 4: Full report + detailed implementation plan + ROI calculation

Output: 20-30 page detailed report

Example 4: Focus on the intelligent customer service field

Function description: Generate an AI scenario map focused on the intelligent customer service field for a certain enterprise

Execution method: Use specific domain focus mode (customer service field)

  • Phase 1: Enterprise research + customer service field information collection (web-search)
  • Phase 2: Focused customer service business analysis
  • Phase 3: Generate 30 scenarios, of which 20 are in the customer service field (67%)
  • Phase 4: In the report, the customer service field accounts for 60%, and all key scenarios in Part 4 are customer service scenarios

Output: A full report focused on customer service