deepresearch

A comprehensive methodology for conducting rigorous, traceable research projects that produce decision-ready deliverables with evidence-based analysis, structured reasoning, and quality gates.

Core Principles

1. Artifact-driven progress: Research is organized into 5 stages, each producing required deliverables
2. Quality gates: Each stage has explicit quality criteria that must be met before proceeding
3. Traceable evidence: Every claim must be traceable to sources with proper citation
4. Structured analysis: Use structured analytic techniques to mitigate bias and improve rigor
5. Uncertainty expression: Clearly distinguish facts, judgments, and speculation with likelihood and confidence
6. Decision-ready outputs: Deliverables are structured for immediate use by decision-makers

Recommended Tools

Cursor IDE Browser (cursor-ide-browser)

The cursor-ide-browser MCP server is highly recommended for DeepResearch projects. It enables browser automation directly within Cursor IDE, making information collection, verification, and OSINT work more efficient and traceable.

Key Use Cases:

1. Information Collection (Stage 3):

   • Navigate to official sources, company websites, regulatory filings
   • Capture screenshots with timestamps for audit trail
   • Extract structured data from web pages
   • Archive web pages before they change or disappear

2. OSINT Verification (Stage 4):

   • Reverse image search on multiple platforms (Google Images, TinEye, Yandex)
   • Verify social media posts and UGC authenticity
   • Check geolocation using map services
   • Capture verification evidence (screenshots, page snapshots)

3. Source Archiving:

   • Take full-page screenshots of key sources
   • Capture page snapshots with accessibility tree for later analysis
   • Document page state at time of access (for reproducibility)

4. Cross-Platform Verification:

   • Navigate between multiple sources to verify consistency
   • Check multiple language versions of same content
   • Verify across different platforms (official site, news, social media)

Best Practices:

• Always capture screenshots/snapshots when accessing sources (for audit trail)
• Use browser navigation to verify links are still active
• Take snapshots before archiving (captures full page state)
• Use browser console to check for dynamic content or hidden information

Integration with Workflow:

• collection-strategist: Use browser to access and archive sources during collection
• verification-expert: Use browser for reverse image search, geolocation verification, UGC checking
• evidence-librarian: Use browser to verify citations and capture source snapshots

See OSINT_VERIFICATION.md for detailed browser-based verification techniques.

Research Stages

Stage A: Task Contract (0→1)

Deliverables:

• Task Contract: Goal, audience, time window, scope, non-goals, deliverable format
• KIQs (Key Intelligence Questions): 3-7 must-answer questions
• Success criteria: Definition of "good enough" and "unobtainable"

Quality Gate 1: Research questions are answerable, falsifiable, with clear time windows. Non-goals are explicit. KIQs ≤ 7 and actionable.

Stage B: Decomposition & Planning (1→Plan)

Deliverables:

• Issue tree / Hypothesis set (MECE decomposition + initial hypotheses)
• Collection plan: Source map, retrieval routes, priorities, verification strategy
• Risk log: Data gaps, timeliness, compliance boundaries, conflicting evidence expectations

Quality Gate 2: Issue tree is MECE. Each sub-question has evidence requirements and source routes. Cross-validation design exists (at least two complementary source types).

Stage C: Collection & Registration (Plan→Evidence)

Deliverables:

• Source Register: Type, time, reliability, bias risk, usable scope for each source
• Evidence Table: Claim→Evidence→Strength→Conflict→Notes
• Collection log: Query strings, timestamps, exclusion reasons (for reproducibility and audit)

Quality Gate 3: Key claims coverage reaches threshold (e.g., 70% of core claims have usable evidence). Evidence is from traceable sources with timestamps/versions. Conflicts are explicitly recorded.

Stage D: Analysis & Convergence (Evidence→Judgment)

Deliverables:

• Structured analysis workbook: ACH/Key Assumptions Check/Red Team (at least 1-2 techniques)
• Key Judgments (3-7): Each with likelihood + confidence + evidence anchors
• Alternative explanations and flip conditions: What would change my mind / signposts

Quality Gate 4: At least 1 alternative explanation exists and is evaluated. Key assumptions are explicit. Most vulnerable assumption is identified. Confidence matches evidence strength.

Stage E: Delivery & Review (Judgment→Product)

Deliverables:

• Deliverable (brief/memo/table/appendices package)
• QA checklist record: Fact-checking, traceable citations, consistent uncertainty expression
• Follow-up actions: Gap list, next collection suggestions, monitoring indicators

Quality Gate 5: Key Judgments are conclusion-first, clear language, audience-appropriate. Each judgment has traceable citations. Uncertainty is expressed consistently. Inference is distinguished from fact.

Quick Start Workflow

⚠️ MANDATORY FIRST STEP: Before starting any DeepResearch project, create AGENTS.md in project root using @cursor-agents-md. This file defines project-specific instructions that all research work must follow.

1. Create AGENTS.md (MANDATORY):

   @cursor-agents-md Create an AGENTS.md file for this DeepResearch project
   

   • Must include project-specific research guidelines
   • Must include reminder to read cursor-agents-md before updates
   • Must be in project root directory

2. Create Task Contract: Define research question, KIQs, scope, and success criteria

3. Build Issue Tree: MECE decomposition with initial hypotheses

4. Develop Collection Plan: Source map with cross-validation strategy

5. Collect & Register: Build Source Register and Evidence Table as you collect

6. Analyze: Apply structured analytic techniques (see STRUCTURED_ANALYSIS.md)

7. Synthesize: Generate Key Judgments with likelihood and confidence

8. Deliver: Create deliverable following REPORT_TEMPLATE.md and run QA_CHECKLIST.md

Evidence-Chain Production Line (12-Stage Workflow)

The complete workflow from problem to usable conclusions, with subagent assignments:

Stage 0: Task Contract Definition

Subagent: research-lead

| Input | Actions | Output | Gate | |-------|---------|--------|------| | Decision-maker's vague question | Define decision goal, boundaries, time window, success criteria | Task Contract v1 | Does "falsifiable judgment" exist? |

⚠️ MANDATORY PREREQUISITE: Before starting Stage 0, ensure AGENTS.md exists in project root. If not, create it using @cursor-agents-md.

Checklist:

• [ ] AGENTS.md exists (created using @cursor-agents-md)
• [ ] Research question is falsifiable and testable
• [ ] Non-goals are explicit
• [ ] Time window defined
• [ ] Success criteria clear
• [ ] Deliverable format specified

Handoff to: methodologist (for KIQ decomposition)

---

Stage 1: KIQ & Claim Draft

Subagents: research-lead + methodologist

| Input | Actions | Output | Gate | |-------|---------|--------|------| | Task Contract | Break into KIQs; form initial Key Claims | Claim Tree v1 | Do "falsifiable" claims exist? |

Collaboration:

• research-lead: Defines KIQs and priority
• methodologist: Ensures claims are falsifiable, suggests hypothesis structure

Handoff to: collection-strategist (for source map design)

Checklist:

• [ ] KIQs ≤ 7
• [ ] Each KIQ has corresponding claims
• [ ] Claims are falsifiable
• [ ] Priority order established

---

Stage 2: Source Map Design

Subagent: collection-strategist

| Input | Actions | Output | Gate | |-------|---------|--------|------| | Claim Tree | Design source type combinations for each claim | Source Map | Each claim has ≥2 independent source types? |

Handoff to: Collectors (Stage 3) + verification-expert (for verification strategy)

Checklist:

• [ ] Each claim mapped to ≥2 source types
• [ ] Cross-validation design exists
• [ ] Retrieval routes specified
• [ ] Archiving strategy defined
• [ ] Language variants considered

---

Stage 3: Collection & Archiving

Subagents: collection-strategist + Collectors

| Input | Actions | Output | Gate | |-------|---------|--------|------| | Source Map | Search, download, archive, tag | Evidence Pool | Is it auditable? Are original copies available? |

Collaboration:

• collection-strategist: Monitors coverage, adjusts strategy
• Collectors: Execute retrieval, archive materials

Recommended Tool: Use cursor-ide-browser to:

• Navigate to sources and capture screenshots/snapshots immediately
• Archive web pages before they change or disappear
• Verify links are still active
• Extract structured data from web pages

Handoff to: verification-expert (Stage 4)

Checklist:

• [ ] Original sources archived
• [ ] Screenshots/snapshots captured (browser evidence)
• [ ] Timestamps recorded
• [ ] Archive links created
• [ ] Metadata captured
• [ ] Coverage threshold met (e.g., 70% of core claims)

---

Stage 4: OSINT Verification

Subagent: verification-expert

| Input | Actions | Output | Gate | |-------|---------|--------|------| | Evidence Pool | Deception removal, time-geography consistency check | Verified Evidence | Does it pass "falsify first" test? |

Recommended Tool: Use cursor-ide-browser to:

• Perform reverse image/video search across multiple platforms
• Verify geolocation using map services (Google Maps/Earth)
• Check social media accounts and capture snapshots
• Verify chronolocation using weather/timezone services
• Capture verification evidence (screenshots, snapshots) at each step

Handoff to: evidence-librarian (Stage 5)

Checklist:

• [ ] UGC verified (source, time, location, originality)
• [ ] Images/videos geolocated (if applicable)
• [ ] Chronolocation verified
• [ ] Consistency checks passed
• [ ] Verification log complete
• [ ] Browser-captured evidence included (screenshots, snapshots)

---

Stage 5: Evidence Registration

Subagent: evidence-librarian

| Input | Actions | Output | Gate | |-------|---------|--------|------| | Verified Evidence | Build Claim–Evidence Table | Evidence Register | Does each key judgment have evidence? |

Handoff to: methodologist (Stage 6) + quant-analyst (for data consistency)

Checklist:

• [ ] Source Register complete
• [ ] Evidence Table built
• [ ] Each claim linked to evidence
• [ ] Conflicts explicitly recorded
• [ ] Citations traceable

---

Stage 6: Structured Reasoning

Subagent: methodologist

| Input | Actions | Output | Gate | |-------|---------|--------|------| | Evidence Register | ACH, hypothesis competition, discriminating evidence | Hypothesis Matrix | Do competing worlds exist? |

Collaboration:

• May consult domain-expert for mechanism plausibility
• May consult quant-analyst for data consistency

Handoff to: quant-analyst (Stage 7) + devils-advocate (Stage 8)

Checklist:

• [ ] ACH matrix complete
• [ ] At least 2 competing hypotheses
• [ ] Discriminating evidence identified
• [ ] Key assumptions checked
• [ ] Alternative explanations evaluated

---

Stage 7: Data Consistency & Sensitivity

Subagent: quant-analyst

| Input | Actions | Output | Gate | |-------|---------|--------|------| | Hypothesis Matrix | Metric unification, sensitivity analysis | Consistency Pack | Are conclusions sensitive to assumptions? |

Handoff to: devils-advocate (Stage 8) + domain-expert (Stage 9)

Checklist:

• [ ] Metrics unified
• [ ] Consistency checks passed
• [ ] Sensitivity analysis complete
• [ ] Error ranges specified
• [ ] Timeline closed (if applicable)

---

Stage 8: Counter-World Attack

Subagent: devils-advocate

| Input | Actions | Output | Gate | |-------|---------|--------|------| | Consistency Pack | Construct counter-worlds, Kill Points | Adversarial Review | Do single-point failures exist? |

Key Activities:

1. Construct 2-4 counterfactual worlds
2. Identify Kill Points (evidence that if falsified, conclusion fails)
3. Create Fragility Map (which judgments sensitive to which assumptions)
4. Write Adversarial Review Memo
5. Design Decision-Failure Simulation

Checklist:

• [ ] At least 2 counterfactual worlds
• [ ] Kill points identified
• [ ] Fragility map complete
• [ ] Adversarial review challenges main conclusion
• [ ] Failure scenarios designed

Handoff to: domain-expert (Stage 9) + methodologist (if re-analysis needed)

---

Stage 9: Domain Mechanism Validation

Subagent: domain-expert

| Input | Actions | Output | Gate | |-------|---------|--------|------| | Adversarial Review | Mechanism plausibility check | Mechanism Memo | Does it violate industry common sense? |

Handoff to: editor (Stage 10)

Checklist:

• [ ] Mechanisms are plausible
• [ ] Industry patterns respected
• [ ] Anomalies flagged
• [ ] Context provided
• [ ] Common sense boundaries checked

---

Stage 10: Conclusion Packaging

Subagent: editor

| Input | Actions | Output | Gate | |-------|---------|--------|------| | Mechanism Memo | Pyramid structure, risk grading | Draft Report | Can conclusions be grasped in 3 minutes? |

Key Activities:

1. Structure: Conclusion-first pyramid
2. Language: Precise uncertainty expression
3. Format: Scannable, forwardable
4. Risk narrative: Clear and actionable

Handoff to: qa-gatekeeper (Stage 11)

Checklist:

• [ ] Conclusion-first structure
• [ ] Key Judgments clear
• [ ] Uncertainty expressed consistently
• [ ] 3-minute grasp test passed
• [ ] Evidence anchors present

---

Stage 11: QA Gate

Subagent: qa-gatekeeper

| Input | Actions | Output | Gate | |-------|---------|--------|------| | Draft Report | Method audit, compliance check | Go / No-Go | Is release permitted? |

Key Activities:

1. Tradecraft QA (all 8 dimensions from rubric)
2. Compliance check (ethics, privacy, permissions)
3. Risk assessment
4. Final Go/No-Go decision

Checklist:

• [ ] All quality gates passed
• [ ] Compliance verified
• [ ] Ethics boundaries respected
• [ ] Risk acceptable
• [ ] Ready for delivery

If No-Go: Return to appropriate stage with specific feedback

Key Resources

Templates & Checklists

• PROJECT_MANAGEMENT.md: Complete project management template with all stages, gates, and tracking tables
• REPORT_TEMPLATE.md: Standard report template with all required sections
• QA_CHECKLIST.md: Pre-delivery quality assurance checklist

Methodology Guides

• STRUCTURED_ANALYSIS.md: Structured analytic techniques (ACH, Key Assumptions Check, Red Team, etc.)
• OSINT_VERIFICATION.md: OSINT verification techniques for UGC, images, videos, geolocation
• UNCERTAINTY_EXPRESSION.md: How to express likelihood, confidence, and distinguish facts/judgments/speculation

Quality Standards

• RUBRIC.md: Evaluation rubric with 8 dimensions and scoring criteria
• ETHICS_GUARDRAILS.md: Compliance and ethics boundaries

Team & Subagents System

All subagent system documentation is integrated into this SKILL.md file. See sections:

• Subagents System Setup: Initialization and setup instructions
• Evidence-Chain Production Line: Complete 12-stage workflow with subagent assignments
• Weekly Research Rituals: Fixed weekly ceremonies
• Subagent Role Definitions: Detailed role descriptions for all 10 subagents

Evidence & Citation Standards

Source Register Minimum Fields

• Type (official/company/media/academic/UGC/database)
• Provenance (who produced, when, version)
• Access path (how obtained/paid/scraped)
• Bias risks (stakeholder interests, propaganda tendency, method limitations)
• Reliability (High/Medium/Low + rationale)
• Use limits (what it can prove)

Evidence Table Structure

• Claim (falsifiable assertion)
• Evidence (citation + summary)
• Supports/Contradicts (which hypothesis)
• Strength (Strong/Medium/Weak: based on method and independence)
• Alternative explanations
• Notes (gaps, next verification steps)

Citation Requirements

• Each Key Judgment: At least 2 independent sources (or 1 primary authoritative + explanation why sufficient)
• Each key number/timeline node: Must be traceable to original source or clear derivation chain
• Conflicting evidence: Must be explicitly presented with explanation of choice and remaining uncertainty

Structured Analytic Techniques

Select technique based on problem type:

• Causal attribution / Who did it → ACH + prioritize disconfirming evidence
• Future prediction / Risk → Scenario planning + indicator framework
• Strong consensus → Devil's Advocacy / Team A-Team B
• Unstable key premises → Key Assumptions Check
• Adversary intent/behavior → Red Team (avoid mirror imaging)

See STRUCTURED_ANALYSIS.md for detailed procedures.

Uncertainty Expression

Three Categories (Kent)

• Fact: Observable, verifiable with high certainty
• Judgment/Estimate: Evidence sufficient but still probabilistic
• Inference/Speculation: Limited evidence, more logic-based

Probability Words (5-tier)

• Almost impossible
• Unlikely
• Possible
• Likely
• Almost certain

Confidence Levels

• High: Multiple independent sources, consistent, high-quality methods
• Medium: Some independent verification, partial consistency
• Low: Single source, high uncertainty, limited verification

Each Key Judgment must include: Likelihood (probability word) + Confidence (High/Medium/Low) + Why (evidence and method rationale)

See UNCERTAINTY_EXPRESSION.md for detailed guidance.

Common Failure Modes

Scope Creep

Symptom: Delivery date approaching but questions multiplying, conclusions becoming vague Solution: Force return to Task Contract. New questions must answer:

1. Will not doing it affect the decision?
2. Is there an evidence path? If unobtainable, move to "future work", not current scope

Last-Minute Citation & Verification

Symptom: Report finished but evidence doesn't match/links broken/inconsistent metrics Solution: Front-load citation and evidence registration:

• Register sources during collection (Source Register)
• Pull evidence from Evidence Table when writing conclusions (not from memory)

Project Rhythm (1-2 week research)

• Day 1: Task Contract + Issue tree + Collection plan (Gate 1/2)
• Day 2-4: Collection + registration + initial Evidence Table (daily Gate 3)
• Day 5: Structured analysis (ACH/Key Assumptions Check) + initial Key Judgments (Gate 4)
• Day 6: Fill gaps, handle conflicts, update confidence
• Day 7: Deliverable writing + Red Team + QA (Gate 5)

Daily Standup (10 minutes)

• What new "usable evidence" was added yesterday (not "what was read")
• What hypothesis/gap will be verified today
• Blockers: Can't get data? Conflicting evidence? Scope change?

Tradecraft Review (every 2-3 days)

• Are claims covered? Are conflicts recorded?
• Is there a tendency toward "feeling-based convergence"?
• Does collection strategy need adjustment (change source types, languages, timeline)?

Red Team / Devil's Advocate (24 hours before delivery)

• What is the most vulnerable point of your conclusion?
• Which evidence is weakest? What if it's wrong?
• Have you "missed alternative explanations"?

Weekly Research Rituals

Fixed weekly ceremonies that support the evidence-chain production line:

Ritual Overview

| Ritual | Frequency | Duration | Participants | Purpose | |--------|-----------|----------|--------------|---------| | Claim Review | Weekly | 30 min | research-lead, methodologist | Prevent scope drift | | Kill Point Review | Weekly | 30 min | devils-advocate, research-lead | Identify single-point failures | | Conflict Evidence Stand-up | Weekly | 15 min | evidence-librarian, all | Make conflicting evidence explicit | | Devil's Day | Bi-weekly | 2 hours | devils-advocate, all | Counter-world attack session | | QA Pre-Gate | Before delivery | 1 hour | qa-gatekeeper, editor | Pre-release failure simulation |

---

1. Claim Review (Monday, 30 min)

Participants: research-lead (facilitator), methodologist
Optional: collection-strategist, domain-expert

Agenda:

1. Review Current Claims (10 min)

   • List all active claims from Claim Tree
   • Check: Are they still falsifiable?
   • Check: Do they still answer KIQs?

2. Scope Check (10 min)

   • Compare claims to Task Contract
   • Identify scope drift
   • Decide: Keep, modify, or remove claims

3. Priority Update (10 min)

   • Re-rank claims by decision impact
   • Identify which claims need evidence first
   • Update collection priorities

Outputs:

• Updated Claim Tree
• Priority Matrix
• Scope Change Log (if any)

Success Criteria:

• [ ] All claims traceable to KIQs
• [ ] No scope drift beyond Task Contract
• [ ] Priorities reflect decision needs

---

2. Kill Point Review (Wednesday, 30 min)

Participants: devils-advocate (facilitator), research-lead, methodologist, evidence-librarian

Agenda:

1. Identify Kill Points (15 min)

   • Review current Key Judgments
   • For each judgment: What evidence, if falsified, would kill it?
   • List all Kill Points

2. Assess Fragility (10 min)

   • Which judgments have single-point failures?
   • Which assumptions are most vulnerable?
   • Create Fragility Map

3. Action Plan (5 min)

   • Which Kill Points need additional evidence?
   • Which need re-verification?
   • Assign follow-up tasks

Outputs:

• Kill-Point List (updated)
• Fragility Map
• Action Items

Success Criteria:

• [ ] All Key Judgments have identified Kill Points
• [ ] Single-point failures flagged
• [ ] Action plan for strengthening weak points

---

3. Conflict Evidence Stand-up (Friday, 15 min)

Participants: evidence-librarian (facilitator), all subagents (brief check-in)

Agenda:

1. New Conflicts (5 min)

   • evidence-librarian reports new conflicting evidence
   • Brief description: What conflicts, why

2. Status Update (5 min)

   • Each subagent: Any conflicts discovered in their work?
   • Quick round: "I found X conflicting with Y"

3. Next Steps (5 min)

   • Which conflicts need investigation?
   • Assign to appropriate subagent

<!-- Content truncated for initial SEO render. Open the source file tab for the full file. -->