github-deep-research
Verifiedby bytedance · Updated Apr 16, 2026
Conduct multi-round deep research on any GitHub Repo. Use when users request comprehensive analysis, timeline reconstruction, competitive analysis, or in-depth investigation of GitHub. Produces structured markdown reports with executive summaries, chronological timelines, metrics analysis, and Mermaid diagrams. Triggers on Github repository URL or open source projects.
See It In Action
I'll kick off deep research on LangGraph to get you a solid due diligence readout.
Starting the research. I'll run multiple rounds of data gathering.
Round 1: GitHub API data collection
Let me find the script and also start gathering data via gh CLI and web searches in parallel.
Good data from Round 1 and 2. Let me do deeper investigation on commercial model, adoption, and red flags.
Let me grab the commit velocity and the security vulnerability details.
Now I have all the data. Let me compile the report.
Done. The full report is saved at research_langgraph_20260413.md. Here's the quick takeaway for your call:
LangGraph in 30 seconds:
- What: Graph-based agent orchestration framework from LangChain Inc. ($1.25B unicorn, $260M raised, backed by Sequoia/Benchmark/IVP)
- Traction: 29K GitHub stars, MIT-licensed, ~600-800 companies in production (LinkedIn, Uber, Elastic, JPMorgan, Klarna)
- Strengths: Best-in-class stateful orchestration, human-in-the-loop, model-agnostic, most mature open-source agent framework
- Watch-outs:
- Security: 4 CVEs in the past year including a critical RCE (all patched, but pattern warrants scrutiny)
- Lock-in: Core is MIT, but commercial LangSmith/Platform create gravitational pull
- Complexity: Steep learning curve; overkill for simple agent patterns where direct provider SDKs suffice
- Contributor concentration: Top 3 contributors are all LangChain employees
Bottom line: Strong choice for complex stateful agents, but evaluate whether your use cases actually need the graph abstraction vs. simpler alternatives. If you proceed, pin versions, self-host where possible, and have an exit strategy.
External Tools
| Tool | Type |
|---|---|
| github | mcp |
| python3 | binary |
Permissions
| Scope | Description |
|---|---|
| filesystem:write | |
| network:outbound | |
| process:spawn |
SKILL.md
GitHub Deep Research Skill
Multi-round research combining GitHub API, web_search, web_fetch to produce comprehensive markdown reports.
Research Workflow
- Round 1: GitHub API
- Round 2: Discovery
- Round 3: Deep Investigation
- Round 4: Deep Dive
Core Methodology
Query Strategy
Broad to Narrow: Start with GitHub API, then general queries, refine based on findings.
Round 1: GitHub API
Round 2: "{topic} overview"
Round 3: "{topic} architecture", "{topic} vs alternatives"
Round 4: "{topic} issues", "{topic} roadmap", "site:github.com {topic}"
Source Prioritization:
- Official docs/repos (highest weight)
- Technical blogs (Medium, Dev.to)
- News articles (verified outlets)
- Community discussions (Reddit, HN)
- Social media (lowest weight, for sentiment)
Research Rounds
Round 1 - GitHub API
Directly execute scripts/github_api.py without read_file():
python /path/to/skill/scripts/github_api.py <owner> <repo> summary
python /path/to/skill/scripts/github_api.py <owner> <repo> readme
python /path/to/skill/scripts/github_api.py <owner> <repo> tree
Available commands (the last argument of github_api.py):
- summary
- info
- readme
- tree
- languages
- contributors
- commits
- issues
- prs
- releases
Round 2 - Discovery (3-5 web_search)
- Get overview and identify key terms
- Find official website/repo
- Identify main players/competitors
Round 3 - Deep Investigation (5-10 web_search + web_fetch)
- Technical architecture details
- Timeline of key events
- Community sentiment
- Use web_fetch on valuable URLs for full content
Round 4 - Deep Dive
- Analyze commit history for timeline
- Review issues/PRs for feature evolution
- Check contributor activity
Report Structure
Follow template in assets/report_template.md:
- Metadata Block - Date, confidence level, subject
- Executive Summary - 2-3 sentence overview with key metrics
- Chronological Timeline - Phased breakdown with dates
- Key Analysis Sections - Topic-specific deep dives
- Metrics & Comparisons - Tables, growth charts
- Strengths & Weaknesses - Balanced assessment
- Sources - Categorized references
- Confidence Assessment - Claims by confidence level
- Methodology - Research approach used
Mermaid Diagrams
Include diagrams where helpful:
Timeline (Gantt):
gantt
title Project Timeline
dateFormat YYYY-MM-DD
section Phase 1
Development :2025-01-01, 2025-03-01
section Phase 2
Launch :2025-03-01, 2025-04-01
Architecture (Flowchart):
flowchart TD
A[User] --> B[Coordinator]
B --> C[Planner]
C --> D[Research Team]
D --> E[Reporter]
Comparison (Pie/Bar):
pie title Market Share
"Project A" : 45
"Project B" : 30
"Others" : 25
Confidence Scoring
Assign confidence based on source quality:
| Confidence | Criteria |
|---|---|
| High (90%+) | Official docs, GitHub data, multiple corroborating sources |
| Medium (70-89%) | Single reliable source, recent articles |
| Low (50-69%) | Social media, unverified claims, outdated info |
Output
Save report as: research_{topic}_{YYYYMMDD}.md
Formatting Rules
- Chinese content: Use full-width punctuation(,。:;!?)
- Technical terms: Provide Wiki/doc URL on first mention
- Tables: Use for metrics, comparisons
- Code blocks: For technical examples
- Mermaid: For architecture, timelines, flows
Best Practices
- Start with official sources - Repo, docs, company blog
- Verify dates from commits/PRs - More reliable than articles
- Triangulate claims - 2+ independent sources
- Note conflicting info - Don't hide contradictions
- Distinguish fact vs opinion - Label speculation clearly
- CRITICAL: Always include inline citations - Use
[citation:Title](URL)format immediately after each claim from external sources - Extract URLs from search results - web_search returns {title, url, snippet} - always use the URL field
- Update as you go - Don't wait until end to synthesize
Citation Examples
Good - With inline citations:
The project gained 10,000 stars within 3 months of launch [citation:GitHub Stats](https://github.com/owner/repo).
The architecture uses LangGraph for workflow orchestration [citation:LangGraph Docs](https://langchain.com/langgraph).
Bad - Without citations:
The project gained 10,000 stars within 3 months of launch.
The architecture uses LangGraph for workflow orchestration.
FAQ
What does github-deep-research do?
Conduct multi-round deep research on any GitHub Repo. Use when users request comprehensive analysis, timeline reconstruction, competitive analysis, or in-depth investigation of GitHub. Produces structured markdown reports with executive summaries, chronological timelines, metrics analysis, and Mermaid diagrams. Triggers on Github repository URL or open source projects.
When should I use github-deep-research?
Use it when you need a repeatable workflow that produces text report.
What does github-deep-research output?
In the evaluated run it produced text report.
How do I install or invoke github-deep-research?
Ask the agent to use this skill when the task matches its documented workflow.
Which agents does github-deep-research support?
Agent support is inferred from the source, but not explicitly declared.
What tools, channels, or permissions does github-deep-research need?
It uses github, python3; channels commonly include text; permissions include filesystem:write, network:outbound, process:spawn.
Is github-deep-research safe to install?
Static analysis marked this skill as medium risk; review side effects and permissions before enabling it.
How is github-deep-research different from an MCP or plugin?
A skill packages instructions and workflow conventions; tools, MCP servers, and plugins are dependencies the skill may call during execution.
Does github-deep-research outperform not using a skill?
About github-deep-research
When to use github-deep-research
You need a comprehensive investigation of a GitHub repository or open source project. You want a timeline, metrics summary, and competitive or architectural analysis of a repo. You need a citation-backed markdown research report about a GitHub project.
When github-deep-research is not the right choice
You only need a quick repo summary or basic file inspection. You need to analyze non-GitHub sources without repository-centric research.
What it produces
Produces text report.