Deep Research
Produce citation-backed deep research drafts with a P6 counter-review team that validates claims, sources, recency, and contradictions.
Overview
Deep Research is an agent skill most often used in Idea (also Validate, Ship) that produces evidence-mapped research drafts and runs a four-reviewer counter-review team before delivery.
Install
npx skills add https://github.com/daymade/claude-code-skills --skill deep-researchWhat is this skill?
- Completeness Review Checklist: structure, citations, evidence tiers, and final consistency gates
- Counter-Review Team with four parallel reviewers plus a coordinator
- claim-validator, source-diversity-checker, recency-validator, and contradiction-finder roles
- Tier A/B requirements for numeric claims and explicit conflict notation
- Deep Research V6 P6 phase dedicated to multi-dimensional draft review
- 4 parallel counter-review agents plus coordinator in P6
- Completeness checklist spans structure, evidence tiers, and final consistency sections
Adoption & trust: 683 installs on skills.sh; 1.2k GitHub stars; 2/3 security scanners passed (skills.sh audits).
What problem does it solve?
You have a research draft full of claims but no systematic way to verify citations, recency, source diversity, or contradictions.
Who is it for?
Solo builders writing decision memos, competitive landscapes, or technical surveys where unsupported claims would be costly.
Skip if: Five-minute fact checks, purely creative writing without evidence requirements, or workflows that cannot spawn a review agent team.
When should I use this skill?
You need Deep Research V6-style drafts with mandatory citations, evidence tables, and P6 counter-review before delivery.
What do I get? / Deliverables
You deliver a checklist-compliant report with mapped evidence, documented limitations, and parallel reviewer reports from the P6 counter-review team.
- Citation-backed research draft
- Claim Validation, Source Diversity, Recency, and Contradiction reports
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Canonical shelf is Idea research because the workflow starts from open questions, but review gates and evidence tables also support validate and ship-quality diligence. Research subphase fits structured questions, evidence tables, tiered citations, and final completeness review before recommendations.
Where it fits
Draft a competitive landscape where each finding links to tiered sources before you pick a product direction.
Stress-test MVP assumptions with recency and contradiction review on market-size or regulatory claims.
Run P6 counter-review on a launch or security memo so public statements match verified evidence.
Prepare long-form thought leadership with source-diversity checks to avoid single-vendor bias.
How it compares
Evidence-gated research pipeline with parallel review agents, not a single-pass 'search the web' prompt.
Common Questions / FAQ
Who is deep-research for?
Solo builders and small teams who need publication-grade research discipline inside their coding agent, especially with Claude Code-style skills.
When should I use deep-research?
At Idea research for landscape studies; at Validate scope for evidence-backed decisions; before Ship launch narratives when claims must be defensible—whenever citations and review gates matter.
Is deep-research safe to install?
A gitleaks scan is noted in-repo; confirm current results on the Security Audits panel on this Prism page and restrict network tools as needed.
SKILL.md
READMESKILL.md - Deep Research
Security scan passed Scanned at: 2026-01-25T15:33:59.582023 Tool: gitleaks + pattern-based validation Content hash: 36de97399b27f8a70dc16d74d4abf39d62918f50c6a8f67385463f9b95cc4839 # Completeness Review Checklist Verify the draft meets all requirements before delivery. ## Structure and Format - All required sections present and ordered - Headings match the format contract - Length targets met per section ## Evidence and Citations - Every claim has a citation - Every numeric claim has at least one Tier A or B source - Conflicting sources are explicitly noted - Evidence table maps claims to sources ## Content Quality - Findings answer the research questions - Recommendations are tied to evidence - Limitations and uncertainty are documented - Terminology is consistent ## Final Checks - Dates and time ranges are consistent - No unsupported claims - No duplicate or conflicting statements # Counter-Review Team 使用指南 Deep Research V6 P6 阶段的专用 Agent Team，并行执行多维度审查。 ## Team 架构 ``` counter-review-coordinator (协调者) ├── claim-validator (声明验证器) ├── source-diversity-checker (来源多样性检查器) ├── recency-validator (时效性验证器) └── contradiction-finder (矛盾发现器) ``` ## Agent 职责 | Agent | 职责 | 输出 | |-------|------|------| | **claim-validator** | 验证声明准确性，识别无证据/弱证据声明 | Claim Validation Report | | **source-diversity-checker** | 检查单一来源依赖，source-type 分布 | Source Diversity Report | | **recency-validator** | 验证时敏声明的新鲜度，AS_OF 合规 | Recency Validation Report | | **contradiction-finder** | 发现内部矛盾，缺失的反向观点 | Contradiction and Bias Report | | **counter-review-coordinator** | 整合所有报告，生成最终 P6 报告 | P6 Counter-Review Report | ## 使用流程 ### 1. 准备输入材料 在 P5 (Draft) 完成后，收集以下材料： ``` inputs/ ├── draft_report.md # P5 起草的报告 ├── citation_registry.md # P3 的引用注册表 ├── task-notes/ │ ├── task-a.md # 子代理研究笔记 │ ├── task-b.md │ └── ... └── p0_config.md # P0 配置 (AS_OF 日期, Mode 等) ``` ### 2. 并行分发任务 向 4 个 specialist agent 同时发送任务： ```bash # 向 claim-validator 发送 SendMessage to: claim-validator 输入: draft_report.md + citation_registry.md + task-notes/ 指令: 验证所有声明的证据支持 # 向 source-diversity-checker 发送 SendMessage to: source-diversity-checker 输入: draft_report.md + citation_registry.md 指令: 检查来源多样性和单一来源依赖 # 向 recency-validator 发送 SendMessage to: recency-validator 输入: draft_report.md + citation_registry.md + p0_config.md 指令: 验证时敏声明的新鲜度 # 向 contradiction-finder 发送 SendMessage to: contradiction-finder 输入: draft_report.md + task-notes/ + citation_registry.md 指令: 发现矛盾和缺失的反向观点 ``` ### 3. 协调汇总 等待 4 个 specialist 完成后，发送给 coordinator： ```bash SendMessage to: counter-review-coordinator 输入: - Claim Validation Report - Source Diversity Report - Recency Validation Report - Contradiction and Bias Report 指令: 整合所有报告，生成最终 P6 Counter-Review Report ``` ### 4. 获取最终输出 Coordinator 输出包含： - 问题汇总（必须 ≥3 个） - 关键争议部分（可直接复制到最终报告） - 强制修复清单 - 质量门状态 ## 质量门要求 | 检查项 | 标准模式 | 轻量模式 | 失败处理 | |--------|---------|---------|---------| | 发现问题数 | ≥3 | ≥3 | 重新审查 | | 关键声明单来源 | 0 | 0 | 补充来源或降级 | | 官方来源占比 | ≥30% | ≥20% | 补充官方来源 | | AS_OF 日期完整 | 100% | 100% | 补充日期 | | 核心争议文档化 | 必填 | 必填 | 补充争议部分 | ## 输出示例 ### Coordinator 最终报告结构 ```markdown # P6 Counter-Review Report ## Executive Summary - Total issues found: 7 (critical: 2, high: 3, medium: 2) - Must-fix before publish: 2 - Recommended improvements: 5 ## Critical Issues (Block Publish) | # | Issue | Location | Source | Fix Required | |---|-------|----------|--------|--------------| | 1 | 市场份额声明无来源 | 3.2节 | 无 | 补充来源或删除 | | 2 | 单一社区来源支持收入数据 | 4.1节 | [12] community | 找官方来源替代 | ## 核心争议 / Key Controversies - **争议 1:** 公司声称增长 50% vs 分析师报告增长 30% - 证据强度: official(公司财报) vs academic(第三方研究) - 建议: 并列呈现两种数据，说明差异原因 ## Mandatory Fixes Checklist - [ ] 补充 3.2 节市场份额来源 - [ ] 替换 4.1 节收入数据来源 - [ ] 添加 AS_OF: 2026-04-03 到所有时敏声明 ## Quality Gates Status | Gate | Status | Notes | |------|--------|-------| | P6 ≥3 issues found | ✅ | 发现 7 个问题 | | No critical claim single-sourced | ❌ | 2 个问题待修复 | | A