Skip to content

aiskillstore/marketplace / skills/727474430/deep-search-and-insight-synthesize

deep-search-and-insight-synthesize

Verified

【深度调研专用】当需要系统性调研、多产品对比、深度分析、技术选型、趋势分析时必须使用本技能。触发关键词:调研、对比、分析、评估、选型、趋势、综合报告、多源验证。通过 3 层深度搜索(阅读多个网页)+ 多源交叉验证,提供比普通 WebSearch 更全面、更准确的研究结果和深度洞察。禁止用普通搜索工具替代深度调研任务。

FileKindSize
SKILL.mdskill md7.5 KB
skill-report.jsonconfig11.5 KB

skill-report.json

View on GitHub
{
  "schema_version": "2.0",
  "meta": {
    "generated_at": "2026-07-04T17:14:24.943Z",
    "slug": "727474430-deep-search-and-insight-synthesize",
    "source_url": "https://github.com/727474430/cli-rule-swtich/tree/main/templates/claude/jige/skills/deep-search-and-insight-synthesize",
    "source_ref": "main",
    "model": "codex",
    "analysis_version": "3.0.0",
    "source_type": "community",
    "content_hash": "4e83c58626461d256ba7ab55dfe78bf0d5262f2af8c7f5124a6a438fd09ef3f5",
    "tree_hash": "b239311f3f74e69771e825b56161d01ddef58d251a2c87829029ee8d0c3eace2",
    "provenance": {
      "model_effective": "codex",
      "fallback_chain": [
        {
          "agent": "codex",
          "outcome": "success"
        }
      ],
      "analysis_version": "3.0.0"
    }
  },
  "skill": {
    "name": "deep-search-and-insight-synthesize",
    "description": "Guides AI assistants through structured multi-layer web research, source validation, comparative analysis, and synthesized insight reports.",
    "summary": "Create evidence-based research briefs from multiple web sources.",
    "icon": "📦",
    "version": "1.0.1",
    "author": "727474430",
    "license": "MIT",
    "category": "research",
    "tags": [
      "deep research",
      "source validation",
      "competitive analysis",
      "trend analysis",
      "insight synthesis"
    ],
    "supported_tools": [
      "claude",
      "codex",
      "claude-code"
    ],
    "risk_factors": [
      "external_commands"
    ]
  },
  "security_audit": {
    "risk_level": "low",
    "is_blocked": false,
    "safe_to_publish": true,
    "analysis_status": "ok",
    "summary": "Static command-execution and obfuscation signals were reviewed against SKILL.md. The command finding is a Markdown inline-code false positive, and the entropy finding is explained by readable Chinese text and formatting. A low-severity semantic finding remains because the skill explicitly directs external web search and QuickChart use.",
    "remediation": [
      {
        "issue": "External services are part of the workflow.",
        "suggestion": "Warn users before sending sensitive research topics, queries, or chart data to search providers or QuickChart.",
        "severity": "low"
      }
    ],
    "risk_factor_evidence": [
      {
        "factor": "external_commands",
        "evidence": [
          {
            "file": "SKILL.md",
            "line_start": 74,
            "line_end": 74
          }
        ]
      }
    ],
    "static_findings": [
      {
        "id": "external_commands:SKILL.md:74:ruby-shell-backtick-execution",
        "category": "external_commands",
        "severity": "medium",
        "pattern": "Ruby/shell backtick execution",
        "file": "SKILL.md",
        "line_start": 74,
        "line_end": 74,
        "snippet": "- 使用`Write`工具创建专业的.md文档"
      },
      {
        "id": "obfuscation:SKILL.md:1:heuristic-high-file-entropy-7-26-bits-possible-b",
        "category": "obfuscation",
        "severity": "high",
        "pattern": "[HEURISTIC] High file entropy (7.26 bits) - possible binary/encrypted content",
        "file": "SKILL.md",
        "line_start": 1,
        "line_end": 1,
        "snippet": "File: SKILL.md"
      }
    ],
    "finding_verdicts": [
      {
        "id": "external_commands:SKILL.md:74:ruby-shell-backtick-execution",
        "verdict": "false_positive",
        "confidence": 0.93,
        "reason": "Line 74 contains Markdown inline code around the word Write, not Ruby or shell backtick execution. It describes creating a Markdown document with a tool and does not execute a command."
      },
      {
        "id": "obfuscation:SKILL.md:1:heuristic-high-file-entropy-7-26-bits-possible-b",
        "verdict": "false_positive",
        "confidence": 0.88,
        "reason": "SKILL.md is readable Markdown with Chinese prose, front matter, headings, and bullet lists. No binary, encrypted, or intentionally hidden content was found."
      }
    ],
    "semantic_findings": [
      {
        "title": "External Network Research Directed",
        "description": "The skill instructs the assistant to use web search tools and the QuickChart API. This can disclose user research queries or chart data to external services.",
        "severity": "low",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 121,
            "line_end": 130
          }
        ],
        "confidence": 0.82,
        "confidence_reasoning": "The instructions explicitly mention network search tooling and QuickChart API use. The risk is limited because this behavior matches the skill purpose and no secret collection is requested."
      }
    ],
    "critical_findings": [],
    "high_findings": [],
    "medium_findings": [],
    "low_findings": [
      {
        "title": "External Network Research Directed",
        "description": "The skill instructs the assistant to use web search tools and the QuickChart API. This can disclose user research queries or chart data to external services.",
        "locations": [
          {
            "file": "SKILL.md",
            "line_start": 121,
            "line_end": 130
          }
        ],
        "review_kind": "security",
        "source_category": "semantic",
        "source_severity": "low",
        "confidence": 0.82,
        "confidence_reasoning": "The instructions explicitly mention network search tooling and QuickChart API use. The risk is limited because this behavior matches the skill purpose and no secret collection is requested."
      }
    ],
    "dangerous_patterns": [],
    "files_scanned": 1,
    "total_lines": 165,
    "audit_model": "codex",
    "audited_at": "2026-07-04T17:14:24.943Z",
    "agent_auto_install_policy": "allowed",
    "manual_install_policy": "allowed"
  },
  "content": {
    "user_title": "Research Complex Topics Deeply",
    "value_statement": "Complex research often needs more than one quick search result. This skill guides Claude, Codex, and Claude Code through layered searches, source checks, and structured synthesis.",
    "seo_keywords": [
      "Claude deep research",
      "Codex research skill",
      "Claude Code research workflow",
      "multi source validation",
      "insight synthesis",
      "competitive analysis",
      "technology evaluation",
      "trend research",
      "research report",
      "QuickChart reports"
    ],
    "actual_capabilities": [
      "Breaks broad research questions into searchable subtopics.",
      "Runs three layers of web research for breadth, depth, and verification.",
      "Compares claims across multiple sources before forming conclusions.",
      "Produces concise Markdown summaries for simple research tasks.",
      "Creates longer Markdown reports with references and Mermaid diagrams.",
      "Suggests HTML report layouts with QuickChart visualizations for complex topics."
    ],
    "limitations": [
      "Requires working web search or MCP search tools.",
      "May send research queries or chart data to external services.",
      "Does not guarantee access to paywalled or restricted sources.",
      "Trend forecasts remain analytical judgments, not certain predictions."
    ],
    "use_cases": [
      {
        "title": "Compare Product Options",
        "description": "Research competing products, summarize tradeoffs, and highlight credible evidence for a buying or roadmap decision.",
        "target_user": "Product managers"
      },
      {
        "title": "Evaluate Technical Choices",
        "description": "Collect current evidence about frameworks, tools, or architectures and present a structured recommendation.",
        "target_user": "Engineering leads"
      },
      {
        "title": "Track Market Trends",
        "description": "Gather recent sources, verify claims, and synthesize patterns for a planning brief or executive update.",
        "target_user": "Research analysts"
      }
    ],
    "prompt_templates": [
      {
        "title": "Research a Topic",
        "scenario": "Beginner request for a concise answer.",
        "prompt": "Research the current state of [topic]. Use multiple reliable sources and give me the key findings, uncertainties, and source links."
      },
      {
        "title": "Compare Options",
        "scenario": "Intermediate request for a decision table.",
        "prompt": "Compare [option A], [option B], and [option C] for [use case]. Evaluate cost, strengths, risks, adoption, and best fit."
      },
      {
        "title": "Build a Decision Report",
        "scenario": "Advanced request for a structured document.",
        "prompt": "Create a detailed research report on [decision]. Include search strategy, evidence summary, source comparison, recommendation, and open questions."
      },
      {
        "title": "Synthesize Trends",
        "scenario": "Advanced request for strategic analysis.",
        "prompt": "Analyze trends in [market or technology] over the last [time period]. Validate claims across sources and separate facts from forecasts."
      }
    ],
    "output_examples": [
      {
        "input": "Compare vector databases for a small AI product team.",
        "output": "A concise comparison covering hosted and self-managed options, tradeoffs, cost drivers, evidence quality, and a recommended shortlist."
      },
      {
        "input": "Research the latest enterprise adoption trends for local AI models.",
        "output": "A trend brief with recent source links, verified claims, conflicting viewpoints, adoption drivers, risks, and near-term outlook."
      },
      {
        "input": "Evaluate whether to use QuickChart for public dashboard images.",
        "output": "A decision note covering capabilities, external service considerations, alternatives, data sensitivity, and implementation guidance."
      }
    ],
    "best_practices": [
      "State the research question, audience, and decision criteria before starting.",
      "Ask for source dates and confidence levels when the topic changes quickly.",
      "Review citations before using the report in high-stakes decisions."
    ],
    "anti_patterns": [
      "Do not use it for private data that should not leave your environment.",
      "Do not treat forecasts as facts without checking source assumptions.",
      "Do not request a long report when a direct answer is enough."
    ],
    "faq": [
      {
        "question": "What does this skill do?",
        "answer": "It guides an AI assistant through layered web research, source validation, and structured insight synthesis."
      },
      {
        "question": "Which tools does it support?",
        "answer": "It is designed for Claude, Codex, and Claude Code when search and file-writing tools are available."
      },
      {
        "question": "Does it browse the web?",
        "answer": "Yes. The workflow expects web search or MCP search tools for current information."
      },
      {
        "question": "Can it create reports?",
        "answer": "Yes. It can guide concise Markdown answers, longer Markdown documents, and HTML-style report plans."
      },
      {
        "question": "Is it safe for sensitive topics?",
        "answer": "Use caution. Research queries and chart data may be sent to external search or chart services."
      },
      {
        "question": "Does it replace expert review?",
        "answer": "No. It improves research structure, but users should verify sources for legal, medical, financial, or security decisions."
      }
    ]
  },
  "file_structure": [
    {
      "name": "SKILL.md",
      "type": "file",
      "path": "SKILL.md",
      "lines": 165
    }
  ]
}
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 196kUpdated Claude CodeCodexVerifiedInstall