Think Tank Report Writer v1.0
One-stop advisory report: topic to draft
Description
🎯 One sentence policy area, produce a directly submittable advisory report in 30 minutes. It's not about 'letting AI write for you,' but packaging 20 years of think tank writing methodology into an engineered pipeline—from topic selection, four-layer penetration, structured drafting to Red-Blue Team quality check, each step goes deeper than you would on your own. ━━━ Who It's For ━━━ ✅ Think tank researchers, government policy researchers, university policy researchers ✅ Writers of internal reference reports, special reports, and policy recommendation reports ✅ Experienced authors who teach advisory report writing to students/teams ━━━ 4 Core Capabilities ━━━ 🎯 Three-Dimensional Topic Scanning Pain point: Topics lack focus, angles are imprecise, recommendations are broad and fragmented. SKILL: Antenna (high-frequency verbs and gaps in policy documents) + Grounding (grassroots pain points) + Side View (horizontal comparison), outputting 3 candidate angles (A1 Problem-Oriented / A2 Comparative Reference / A3 Major Project Proposal each). 🔍 Four-Layer Penetration Analysis Pain point: Reports stay at the surface level, failing to uncover institutional roots, leading to rejection by superiors. SKILL: Phenomenon → Institution → Interests → History, layer by layer penetration. Dig three layers deeper than peers. 📋 Structured Drafting Pain point: Unfamiliar with party-government document formats, confused by three types of internal reference styles. SKILL: A1 Problem-Oriented / A2 Comparative Reference / A3 Major Project Proposal three styles; format 100% aligned with GB/T 9704-2012; out of the box. 🛡️ Red-Blue Team Adversarial Quality Check Pain point: Can't revise your own draft / Obvious AI feel / Don't know where the issues are. SKILL: Switch to reviewer perspective to critique every sentence, graded as P0 (blocks submission) / P1 (recommended fix) / P2 (optional optimization); built-in four-layer de-AI norms (banned words + sentence restructuring + language strategy + structure strategy). ━━━ 📦 One Input, Five Deliverables ━━━ Not just finished writing, but a set of reusable research assets: 1️⃣ Main report docx — 3500-5000 words, party-government document format, directly submittable 2️⃣ Topic angle candidate library — The 2 unused angles already scanned three-dimensionally, ready for next topic 3️⃣ Four-layer penetration analysis framework card — A thinking template transferable to any policy issue 4️⃣ Policy material card library — Data/policy/cases categorized and archived, an asset that grows with use 5️⃣ Red-Blue Team quality check log — Each edit includes [original text + diagnosis + revised text + rationale] ━━━ How It Works ━━━ ⚙️ Five-station pipeline, user confirmation required at each station Station 1: Three-Dimensional Topic Scanning → Station 2: Four-Layer Penetration Analysis → Station 3: Material Dehydration and Transformation → Station 4: Structured Drafting → Station 5: Red-Blue Team Quality Check ⚙️ Dashboard mechanism Each output includes a progress bar (completed / in progress / pending) + a locked parameter list to counter context drift. ⚙️ User confirmation system At the end of each station: ask if you want to proceed. Unsatisfied? You can redo until it's right. ⚙️ Fast track Experienced users with complete information can enter 'fast track' to skip pre-confirmation and go straight to drafting + quality check. ━━━ What We Don't Do (Aesthetic Lockdown) ━━━ ✕ No 'free style' option — the three special report styles have been validated by party-government organs for 20 years; free combining only leads to loss of control ✕ No 'one-click full auto' mode — five-station mandatory confirmation is a quality baseline, not a bug ✕ No delivery without fixing P0 blocking issues — quality relies on system, not luck ━━━ How to Start ━━━ Just say one sentence after opening: 'I want to write an advisory report on county-level elderly care services' 'I want to write an advisory report on financing difficulties of specialized and new enterprises' 'I want to write an advisory report on implementation of grassroots burden reduction policies' ⏱️ Total time about 30 minutes (compared to 8-12 days for manual writing, time reduction of 99%)
Related Skills
View all
ResearchAcademic Paper Writing v4.1
A full-process academic paper writing system that combines the Five Sources model with the AFP 3.1 three-layer architecture. Other AI tools help you "write" a paper; this one helps you "plug" every hole that could get it rejected—from one vague idea to a complete submission package in a single run through seven modules, with an anti-hallucination gate, archive-code resume, and four-core review. [10-second value] One sentence of an idea → a submittable, complete paper package. Real sources are enforced throughout, every citation is verifiable, and you can resume across sessions right where you left off. [Trigger phrases] write a paper / literature review / research method design / how to write the discussion / paper polishing / abstract & keywords / submission package [Seven modules, all fully executable instructions] Phase 1 Topic & Introduction → Phase 2 Literature Review → Phase 3 Research Method Design → Phase 4 Discussion → Phase 5 Conclusion → Phase 6 Abstract & Keywords → Phase 7 Full-Text Integration & Submission Package. Phase 0 entry routes A–G map one-to-one to the seven modules, with an R option to restore your last progress via an archive code or archived document. [v4.1 upgrades] 1. Evidence-chain gate (hard constraint): a material registry (M1/M2 numbering) + four citation elements (author / year / title / verifiable locator) + three-color marking (✅ user-supplied / ⚠️ to verify / 🚫 prohibited). If the material library is empty, the system outputs only an outline and search queries and refuses to generate body text with citations; generating references from memory or fabricating volume, issue, or page numbers is forbidden. 2. Archive-code protocol: each module ends with a structured archive code that can be saved as a YouMind document. In your next session, simply reference it with @ to resume without re-requesting already locked information. 3. Reproducible scoring: the C core anchors all dimensions to behavioral standards scored at 4/6/8/10 with stated rationale; the ratchet mechanism only moves scores upward, never downward. 4. Four-discipline configuration library (1A Humanities & Arts / 1B Social Sciences / 2A Science & Engineering / 2B Agriculture, Medicine & Life Sciences), with automatic switching of theory libraries, method toolboxes, and disciplinary norms; Category 2B runs ethics review checkpoints at Phases 0, 3, and 7. 5. Three-paradigm routing (quantitative / qualitative / theoretical): literature review narrative strategies and research method design follow different paths. 6. Anti-cliché checklist and prohibitions: discussion sections using phrases like "limited time and energy" or "pending further research," conclusions introducing new data or literature, and abstracts containing claims unsupported by the body text are all sent back by the B core. 7. Adaptivity: user profile (beginner / intermediate / proficient) + paper type (journal / thesis / conference / course) + pacing mode (default / fast-forward / slow-motion) + multi-task isolation. 8. Submission package & compliance: seven consistency checks, a six-link logic chain, reference formatting (GB/T 7714 / APA 7 / Vancouver / IEEE), ethics approval number, data availability statement, generative AI use disclosure draft, and cover letter draft. [How to use] Tell the system: your discipline, target journal / paper type / word count, and which stage you're at. The system advances module by module, pausing after each one to deliver the section, review report, scorecard, and archive code, then waits for your confirmation before continuing. [Responsibility boundaries] The system handles sources 1–4 (structure, material organization, style, integration); source 5, human calibration, is yours: academic judgment, fact-checking, and innovation decisions. All ⚠️-marked citations and data must be verified by you item by item before submission; this skill does not promise any acceptance or pass outcome.
HE Document Write&Review v3.0
🎯 Got your application rejected? Can't find the highlight for your proposal? Don't know where to start with review comments? Three real scenarios: 🔸 Scenario 1: Application season anxiety—Reviewer feedback: "Insufficient theoretical support, vague policy basis." Unsure which documents to cite or which theoretical framework to use. 🔸 Scenario 2: Proposal writing dilemma—In charge of course construction plan: objectives, tasks, pathways, evaluation… each part needs writing, but you feel the logic is not rigorous enough and worry about being questioned on "feasibility" during review. 🔸 Scenario 3: Review dilemma—Need to write peer review comments: must point out issues while maintaining professionalism, be well-founded but not too harsh. How to strike the balance? 💡 What can this system do for you? Not just give advice—it writes, revises, and reviews for you directly. 📝 Writing Mode: From topic to final document Enter your topic and existing materials. The system automatically identifies the document type (application/proposal/report/review). Automatically matches authoritative policy documents and theoretical support. Generates content chapter by chapter, each with evidence, logic, and facts. Key promise: Never fabricates data; clearly tells you what's missing. 🔍 Review Mode: Expert-level diagnosis Upload your text. Professional scoring across 7 dimensions (value, alignment, completeness, innovation, feasibility, support, expression quality). Precisely identifies problem areas. Provides specific revision suggestions + example rewrites. Not general advice, but paragraph-level specific guidance. ✏️ Revision Optimization Mode: Precision enhancement Strengthens arguments based on existing text. Optimizes expression, eliminates empty talk and clichés. Standardizes terminology and logic. Improves overall competitiveness. ⚡ Three Core Mechanisms (Unique) 🛡️ Firewall Mechanism Built-in "fact boundary": User-provided real data is never fabricated; policy basis must have sources; theoretical support cannot be misapplied. Every sentence you see can be traced back to its source. 🔄 Multi-core Adversarial Engine One core writes, another specifically checks for errors. Like having a strict auditor watching, ensuring no "unsubstantiated facts," "logic gaps," or "policy mismatches" occur. 📊 Stepwise Guidance Doesn't ask you 20 questions at once—identifies the most critical gaps and asks only the 3–5 most necessary questions. After each stage, clearly tells you "what's done," "what's missing," and "what to do next." 🎯 Scope of Application (All Higher Education Scenarios) ✅ Teaching achievement award applications (institutional/provincial/national) ✅ Quality engineering project applications (top courses/teaching teams/textbooks, etc.) ✅ Course construction plans, major construction plans ✅ Major self-assessment reports, course acceptance reports ✅ Expert review comments, peer reviews ✅ Education reform project applications, closing reports 🚀 User Experience Writing an application from scratch: Provide the topic and basic materials → System identifies document type, takes inventory, matches policies and theories → Generates outline → Writes chapter by chapter → Consolidates → Get a draft in 1 hour. Reviewing existing text: Upload document → System automatically scores → Lists main issues → Provides revision suggestions and example rewrites → Get review report in 20 minutes. Optimizing existing plan: Provide existing text and optimization direction → System diagnoses weaknesses → Strengthens arguments, optimizes expression → Get optimized version in 30 minutes. 👉 Try it now—make higher education document writing no longer a burden. This is not just a writing assistant; it's an intelligent engine that understands higher education rules, review standards, and professional expression.
Grant Proposal Review PRO V2.0
🎯 Core Functionality Overview This is an intelligent review and optimization system specially designed for national social science, education ministry, and provincial grant applications. It simulates the thinking mode of a senior review expert with 15 years of experience, ensuring academic rigor and competitiveness through three core mechanisms. 🔧 Three Core Mechanisms 1️⃣ 12-Step Structured Methodology Covers the full lifecycle of grant proposal review: Phase 1-3: Basic Diagnosis - In-depth analysis of announcement (funding priorities, review criteria, application requirements) - Cross-disciplinary type judgment (precise identification of 8 types) - Research GAP five-dimension identification (theory/methodology/empirical/policy/technology) Phase 4-7: Core Element Review - Research question TMAQ model analysis (theory/methodology/approach/question four dimensions) - Research objective SMART principle test - Research content framework completeness assessment - Research approach type matching (6 types) Phase 8-10: Deep Quality Enhancement - Precise extraction of key difficulties (distinguish criteria + breakthrough paths) - Innovation point seven-dimension mining - Feasibility seven-dimension argumentation Phase 11-12: Overall Optimization - Nine-dimension quality check (academic rigor, innovativeness, feasibility, etc.) - Comprehensive optimization suggestions and final report 2️⃣ Dual-Core Adversarial Mechanism (Builder vs Supervisor) Working Principle: - Builder (academic writer): Generates optimization plans based on user materials - Supervisor (top journal reviewer): Challenges Builder's plans with the strictest standards - Adversarial iteration: 3 rounds of confrontation to ensure plans are robust Application Scenarios: - Innovation point mining: Builder proposes innovation points → Supervisor questions novelty → iterative optimization - Feasibility argumentation: Builder designs plan → Supervisor challenges feasibility → supplementary argumentation - Literature citation: Builder cites literature → Supervisor verifies authenticity → ensure academic standards 3️⃣ Literature Authenticity Verification Mechanism Two working modes: Mode A: Placeholder Mode (Default) - Use markers like [Literature Placeholder-001] in place of specific references - Output a Literature Requirement List specifying search requirements for each placeholder - User searches and fills in real references Mode B: Real-Time Verification Mode - Call Google Scholar to verify literature authenticity in real time - Generate Literature Verification Report (authenticity/relevance/authority scores) - Ensure every citation is traceable Preventing AI Hallucination: - Prohibits fabricating authors, journals, DOIs - All references must be verified or marked as placeholders - Guarantees academic integrity bottom line 💡 Core Value and Applicable Scenarios ✅ Key Pain Points Addressed 1. Academic sloppiness: AI-generated content often includes fake references, logical gaps 2. Insufficient innovation: Difficulty uncovering true academic innovation points 3. Weak feasibility: Research plans lack systematic argumentation 4. Cross-disciplinary difficulty: Interdisciplinary topics often fall between two stools 🎓 Target Users - University faculty (social sciences, education, humanities) - Researchers (applying for national and provincial grants) - Academic teams (needing systematic review processes) 📋 Typical Workflow 1. Input: Upload announcement + proposal draft 2. Review: System executes 12-step structured analysis 3. Adversarial: Dual-core mechanism iteratively optimizes key sections 4. Verification: Literature authenticity check 5. Output: Complete review report + optimization suggestions + literature list 🔍 Differences from Traditional Review | Dimension | Traditional Human Review | Expert Review System | |-----------|------------------------|----------------------| | Review depth | Depends on personal experience | 12-step structured + 9D QC | | Academic rigor | Hard to fully audit | Literature verification + dual-core adversarial | | Innovation mining | Subjective judgment | 7-dimension systematic analysis | | Feasibility argumentation | Experience-driven | 7-dimension item-by-item argumentation | | Consistency | Varies by individual | Standardized process | | Efficiency | Days to weeks | 1-2 hours for initial review | The core advantage of this system is: it makes the tacit knowledge of a 15-year senior review expert explicit, structured, and replicable, enabling every user to receive top-level expert review services.
Think Tank Report Writer v1.0
One-stop advisory report: topic to draft
Description
🎯 One sentence policy area, produce a directly submittable advisory report in 30 minutes. It's not about 'letting AI write for you,' but packaging 20 years of think tank writing methodology into an engineered pipeline—from topic selection, four-layer penetration, structured drafting to Red-Blue Team quality check, each step goes deeper than you would on your own. ━━━ Who It's For ━━━ ✅ Think tank researchers, government policy researchers, university policy researchers ✅ Writers of internal reference reports, special reports, and policy recommendation reports ✅ Experienced authors who teach advisory report writing to students/teams ━━━ 4 Core Capabilities ━━━ 🎯 Three-Dimensional Topic Scanning Pain point: Topics lack focus, angles are imprecise, recommendations are broad and fragmented. SKILL: Antenna (high-frequency verbs and gaps in policy documents) + Grounding (grassroots pain points) + Side View (horizontal comparison), outputting 3 candidate angles (A1 Problem-Oriented / A2 Comparative Reference / A3 Major Project Proposal each). 🔍 Four-Layer Penetration Analysis Pain point: Reports stay at the surface level, failing to uncover institutional roots, leading to rejection by superiors. SKILL: Phenomenon → Institution → Interests → History, layer by layer penetration. Dig three layers deeper than peers. 📋 Structured Drafting Pain point: Unfamiliar with party-government document formats, confused by three types of internal reference styles. SKILL: A1 Problem-Oriented / A2 Comparative Reference / A3 Major Project Proposal three styles; format 100% aligned with GB/T 9704-2012; out of the box. 🛡️ Red-Blue Team Adversarial Quality Check Pain point: Can't revise your own draft / Obvious AI feel / Don't know where the issues are. SKILL: Switch to reviewer perspective to critique every sentence, graded as P0 (blocks submission) / P1 (recommended fix) / P2 (optional optimization); built-in four-layer de-AI norms (banned words + sentence restructuring + language strategy + structure strategy). ━━━ 📦 One Input, Five Deliverables ━━━ Not just finished writing, but a set of reusable research assets: 1️⃣ Main report docx — 3500-5000 words, party-government document format, directly submittable 2️⃣ Topic angle candidate library — The 2 unused angles already scanned three-dimensionally, ready for next topic 3️⃣ Four-layer penetration analysis framework card — A thinking template transferable to any policy issue 4️⃣ Policy material card library — Data/policy/cases categorized and archived, an asset that grows with use 5️⃣ Red-Blue Team quality check log — Each edit includes [original text + diagnosis + revised text + rationale] ━━━ How It Works ━━━ ⚙️ Five-station pipeline, user confirmation required at each station Station 1: Three-Dimensional Topic Scanning → Station 2: Four-Layer Penetration Analysis → Station 3: Material Dehydration and Transformation → Station 4: Structured Drafting → Station 5: Red-Blue Team Quality Check ⚙️ Dashboard mechanism Each output includes a progress bar (completed / in progress / pending) + a locked parameter list to counter context drift. ⚙️ User confirmation system At the end of each station: ask if you want to proceed. Unsatisfied? You can redo until it's right. ⚙️ Fast track Experienced users with complete information can enter 'fast track' to skip pre-confirmation and go straight to drafting + quality check. ━━━ What We Don't Do (Aesthetic Lockdown) ━━━ ✕ No 'free style' option — the three special report styles have been validated by party-government organs for 20 years; free combining only leads to loss of control ✕ No 'one-click full auto' mode — five-station mandatory confirmation is a quality baseline, not a bug ✕ No delivery without fixing P0 blocking issues — quality relies on system, not luck ━━━ How to Start ━━━ Just say one sentence after opening: 'I want to write an advisory report on county-level elderly care services' 'I want to write an advisory report on financing difficulties of specialized and new enterprises' 'I want to write an advisory report on implementation of grassroots burden reduction policies' ⏱️ Total time about 30 minutes (compared to 8-12 days for manual writing, time reduction of 99%)
Related Skills
View allResearchAcademic Paper Writing v4.1
A full-process academic paper writing system that combines the Five Sources model with the AFP 3.1 three-layer architecture. Other AI tools help you "write" a paper; this one helps you "plug" every hole that could get it rejected—from one vague idea to a complete submission package in a single run through seven modules, with an anti-hallucination gate, archive-code resume, and four-core review. [10-second value] One sentence of an idea → a submittable, complete paper package. Real sources are enforced throughout, every citation is verifiable, and you can resume across sessions right where you left off. [Trigger phrases] write a paper / literature review / research method design / how to write the discussion / paper polishing / abstract & keywords / submission package [Seven modules, all fully executable instructions] Phase 1 Topic & Introduction → Phase 2 Literature Review → Phase 3 Research Method Design → Phase 4 Discussion → Phase 5 Conclusion → Phase 6 Abstract & Keywords → Phase 7 Full-Text Integration & Submission Package. Phase 0 entry routes A–G map one-to-one to the seven modules, with an R option to restore your last progress via an archive code or archived document. [v4.1 upgrades] 1. Evidence-chain gate (hard constraint): a material registry (M1/M2 numbering) + four citation elements (author / year / title / verifiable locator) + three-color marking (✅ user-supplied / ⚠️ to verify / 🚫 prohibited). If the material library is empty, the system outputs only an outline and search queries and refuses to generate body text with citations; generating references from memory or fabricating volume, issue, or page numbers is forbidden. 2. Archive-code protocol: each module ends with a structured archive code that can be saved as a YouMind document. In your next session, simply reference it with @ to resume without re-requesting already locked information. 3. Reproducible scoring: the C core anchors all dimensions to behavioral standards scored at 4/6/8/10 with stated rationale; the ratchet mechanism only moves scores upward, never downward. 4. Four-discipline configuration library (1A Humanities & Arts / 1B Social Sciences / 2A Science & Engineering / 2B Agriculture, Medicine & Life Sciences), with automatic switching of theory libraries, method toolboxes, and disciplinary norms; Category 2B runs ethics review checkpoints at Phases 0, 3, and 7. 5. Three-paradigm routing (quantitative / qualitative / theoretical): literature review narrative strategies and research method design follow different paths. 6. Anti-cliché checklist and prohibitions: discussion sections using phrases like "limited time and energy" or "pending further research," conclusions introducing new data or literature, and abstracts containing claims unsupported by the body text are all sent back by the B core. 7. Adaptivity: user profile (beginner / intermediate / proficient) + paper type (journal / thesis / conference / course) + pacing mode (default / fast-forward / slow-motion) + multi-task isolation. 8. Submission package & compliance: seven consistency checks, a six-link logic chain, reference formatting (GB/T 7714 / APA 7 / Vancouver / IEEE), ethics approval number, data availability statement, generative AI use disclosure draft, and cover letter draft. [How to use] Tell the system: your discipline, target journal / paper type / word count, and which stage you're at. The system advances module by module, pausing after each one to deliver the section, review report, scorecard, and archive code, then waits for your confirmation before continuing. [Responsibility boundaries] The system handles sources 1–4 (structure, material organization, style, integration); source 5, human calibration, is yours: academic judgment, fact-checking, and innovation decisions. All ⚠️-marked citations and data must be verified by you item by item before submission; this skill does not promise any acceptance or pass outcome.
HE Document Write&Review v3.0
🎯 Got your application rejected? Can't find the highlight for your proposal? Don't know where to start with review comments? Three real scenarios: 🔸 Scenario 1: Application season anxiety—Reviewer feedback: "Insufficient theoretical support, vague policy basis." Unsure which documents to cite or which theoretical framework to use. 🔸 Scenario 2: Proposal writing dilemma—In charge of course construction plan: objectives, tasks, pathways, evaluation… each part needs writing, but you feel the logic is not rigorous enough and worry about being questioned on "feasibility" during review. 🔸 Scenario 3: Review dilemma—Need to write peer review comments: must point out issues while maintaining professionalism, be well-founded but not too harsh. How to strike the balance? 💡 What can this system do for you? Not just give advice—it writes, revises, and reviews for you directly. 📝 Writing Mode: From topic to final document Enter your topic and existing materials. The system automatically identifies the document type (application/proposal/report/review). Automatically matches authoritative policy documents and theoretical support. Generates content chapter by chapter, each with evidence, logic, and facts. Key promise: Never fabricates data; clearly tells you what's missing. 🔍 Review Mode: Expert-level diagnosis Upload your text. Professional scoring across 7 dimensions (value, alignment, completeness, innovation, feasibility, support, expression quality). Precisely identifies problem areas. Provides specific revision suggestions + example rewrites. Not general advice, but paragraph-level specific guidance. ✏️ Revision Optimization Mode: Precision enhancement Strengthens arguments based on existing text. Optimizes expression, eliminates empty talk and clichés. Standardizes terminology and logic. Improves overall competitiveness. ⚡ Three Core Mechanisms (Unique) 🛡️ Firewall Mechanism Built-in "fact boundary": User-provided real data is never fabricated; policy basis must have sources; theoretical support cannot be misapplied. Every sentence you see can be traced back to its source. 🔄 Multi-core Adversarial Engine One core writes, another specifically checks for errors. Like having a strict auditor watching, ensuring no "unsubstantiated facts," "logic gaps," or "policy mismatches" occur. 📊 Stepwise Guidance Doesn't ask you 20 questions at once—identifies the most critical gaps and asks only the 3–5 most necessary questions. After each stage, clearly tells you "what's done," "what's missing," and "what to do next." 🎯 Scope of Application (All Higher Education Scenarios) ✅ Teaching achievement award applications (institutional/provincial/national) ✅ Quality engineering project applications (top courses/teaching teams/textbooks, etc.) ✅ Course construction plans, major construction plans ✅ Major self-assessment reports, course acceptance reports ✅ Expert review comments, peer reviews ✅ Education reform project applications, closing reports 🚀 User Experience Writing an application from scratch: Provide the topic and basic materials → System identifies document type, takes inventory, matches policies and theories → Generates outline → Writes chapter by chapter → Consolidates → Get a draft in 1 hour. Reviewing existing text: Upload document → System automatically scores → Lists main issues → Provides revision suggestions and example rewrites → Get review report in 20 minutes. Optimizing existing plan: Provide existing text and optimization direction → System diagnoses weaknesses → Strengthens arguments, optimizes expression → Get optimized version in 30 minutes. 👉 Try it now—make higher education document writing no longer a burden. This is not just a writing assistant; it's an intelligent engine that understands higher education rules, review standards, and professional expression.
Grant Proposal Review PRO V2.0
🎯 Core Functionality Overview This is an intelligent review and optimization system specially designed for national social science, education ministry, and provincial grant applications. It simulates the thinking mode of a senior review expert with 15 years of experience, ensuring academic rigor and competitiveness through three core mechanisms. 🔧 Three Core Mechanisms 1️⃣ 12-Step Structured Methodology Covers the full lifecycle of grant proposal review: Phase 1-3: Basic Diagnosis - In-depth analysis of announcement (funding priorities, review criteria, application requirements) - Cross-disciplinary type judgment (precise identification of 8 types) - Research GAP five-dimension identification (theory/methodology/empirical/policy/technology) Phase 4-7: Core Element Review - Research question TMAQ model analysis (theory/methodology/approach/question four dimensions) - Research objective SMART principle test - Research content framework completeness assessment - Research approach type matching (6 types) Phase 8-10: Deep Quality Enhancement - Precise extraction of key difficulties (distinguish criteria + breakthrough paths) - Innovation point seven-dimension mining - Feasibility seven-dimension argumentation Phase 11-12: Overall Optimization - Nine-dimension quality check (academic rigor, innovativeness, feasibility, etc.) - Comprehensive optimization suggestions and final report 2️⃣ Dual-Core Adversarial Mechanism (Builder vs Supervisor) Working Principle: - Builder (academic writer): Generates optimization plans based on user materials - Supervisor (top journal reviewer): Challenges Builder's plans with the strictest standards - Adversarial iteration: 3 rounds of confrontation to ensure plans are robust Application Scenarios: - Innovation point mining: Builder proposes innovation points → Supervisor questions novelty → iterative optimization - Feasibility argumentation: Builder designs plan → Supervisor challenges feasibility → supplementary argumentation - Literature citation: Builder cites literature → Supervisor verifies authenticity → ensure academic standards 3️⃣ Literature Authenticity Verification Mechanism Two working modes: Mode A: Placeholder Mode (Default) - Use markers like [Literature Placeholder-001] in place of specific references - Output a Literature Requirement List specifying search requirements for each placeholder - User searches and fills in real references Mode B: Real-Time Verification Mode - Call Google Scholar to verify literature authenticity in real time - Generate Literature Verification Report (authenticity/relevance/authority scores) - Ensure every citation is traceable Preventing AI Hallucination: - Prohibits fabricating authors, journals, DOIs - All references must be verified or marked as placeholders - Guarantees academic integrity bottom line 💡 Core Value and Applicable Scenarios ✅ Key Pain Points Addressed 1. Academic sloppiness: AI-generated content often includes fake references, logical gaps 2. Insufficient innovation: Difficulty uncovering true academic innovation points 3. Weak feasibility: Research plans lack systematic argumentation 4. Cross-disciplinary difficulty: Interdisciplinary topics often fall between two stools 🎓 Target Users - University faculty (social sciences, education, humanities) - Researchers (applying for national and provincial grants) - Academic teams (needing systematic review processes) 📋 Typical Workflow 1. Input: Upload announcement + proposal draft 2. Review: System executes 12-step structured analysis 3. Adversarial: Dual-core mechanism iteratively optimizes key sections 4. Verification: Literature authenticity check 5. Output: Complete review report + optimization suggestions + literature list 🔍 Differences from Traditional Review | Dimension | Traditional Human Review | Expert Review System | |-----------|------------------------|----------------------| | Review depth | Depends on personal experience | 12-step structured + 9D QC | | Academic rigor | Hard to fully audit | Literature verification + dual-core adversarial | | Innovation mining | Subjective judgment | 7-dimension systematic analysis | | Feasibility argumentation | Experience-driven | 7-dimension item-by-item argumentation | | Consistency | Varies by individual | Standardized process | | Efficiency | Days to weeks | 1-2 hours for initial review | The core advantage of this system is: it makes the tacit knowledge of a 15-year senior review expert explicit, structured, and replicable, enabling every user to receive top-level expert review services.
Find your next favorite skill
Explore more curated AI skills for research, creation, and everyday work.