github-deep-research

Name: github-deep-research
Author: OpsinTech

91stars18forks

How to install github-deep-research?

opsintech/opsintech-platform/github-deep-research

$npx -y skills add OpsinTech/opsintech-platform --skill github-deep-research

Installs into the current project.

Summary

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

Files · 1

View on GitHub

SKILL.md

1	# GitHub Deep Research Skill
2
3	Multi-round research combining GitHub API, web_search, web_fetch to produce comprehensive markdown reports.
4
5	## Research Workflow
6
7	- Round 1: GitHub API
8	- Round 2: Discovery
9	- Round 3: Deep Investigation
10	- Round 4: Deep Dive
11
12	## Core Methodology
13
14	### Query Strategy
15
16	Broad to Narrow: Start with GitHub API, then general queries, refine based on findings.
17
18	```
19	Round 1: GitHub API
20	Round 2: "{topic} overview"
21	Round 3: "{topic} architecture", "{topic} vs alternatives"
22	Round 4: "{topic} issues", "{topic} roadmap", "site:github.com {topic}"
23	```
24
25	Source Prioritization:
26	1. Official docs/repos (highest weight)
27	2. Technical blogs (Medium, Dev.to)
28	3. News articles (verified outlets)
29	4. Community discussions (Reddit, HN)
30	5. Social media (lowest weight, for sentiment)
31
32	### Research Rounds
33
34	Round 1 - GitHub API
35	Directly execute `scripts/github_api.py` without `read_file()`:
36	```bash
37	python /path/to/skill/scripts/github_api.py <owner> <repo> summary
38	python /path/to/skill/scripts/github_api.py <owner> <repo> readme
39	python /path/to/skill/scripts/github_api.py <owner> <repo> tree
40	```
41
42	Available commands (the last argument of `github_api.py`):
43	- summary
44	- info
45	- readme
46	- tree
47	- languages
48	- contributors
49	- commits
50	- issues
51	- prs
52	- releases
53
54	Round 2 - Discovery (3-5 web_search)
55	- Get overview and identify key terms
56	- Find official website/repo
57	- Identify main players/competitors
58
59	Round 3 - Deep Investigation (5-10 web_search + web_fetch)
60	- Technical architecture details
61	- Timeline of key events
62	- Community sentiment
63	- Use web_fetch on valuable URLs for full content
64
65	Round 4 - Deep Dive
66	- Analyze commit history for timeline
67	- Review issues/PRs for feature evolution
68	- Check contributor activity
69
70	## Report Structure
71
72	Follow template in `assets/report_template.md`:
73
74	1. Metadata Block - Date, confidence level, subject
75	2. Executive Summary - 2-3 sentence overview with key metrics
76	3. Chronological Timeline - Phased breakdown with dates
77	4. Key Analysis Sections - Topic-specific deep dives
78	5. Metrics & Comparisons - Tables, growth charts
79	6. Strengths & Weaknesses - Balanced assessment
80	7. Sources - Categorized references
81	8. Confidence Assessment - Claims by confidence level
82	9. Methodology - Research approach used
83
84	### Mermaid Diagrams
85
86	Include diagrams where helpful:
87
88	Timeline (Gantt):
89	```mermaid
90	gantt
91	title Project Timeline
92	dateFormat YYYY-MM-DD
93	section Phase 1
94	Development :2025-01-01, 2025-03-01
95	section Phase 2
96	Launch :2025-03-01, 2025-04-01
97	```
98
99	Architecture (Flowchart):
100	```mermaid
101	flowchart TD
102	A[User] --> B[Coordinator]
103	B --> C[Planner]
104	C --> D[Research Team]
105	D --> E[Reporter]
106	```
107
108	Comparison (Pie/Bar):
109	```mermaid
110	pie title Market Share
111	"Project A" : 45
112	"Project B" : 30
113	"Others" : 25
114	```
115
116	## Confidence Scoring
117
118	Assign confidence based on source quality:
119
120	\| Confidence \| Criteria \|
121	\|------------\|----------\|
122	\| High (90%+) \| Official docs, GitHub data, multiple corroborating sources \|
123	\| Medium (70-89%) \| Single reliable source, recent articles \|
124	\| Low (50-69%) \| Social media, unverified claims, outdated info \|
125
126	## Output
127
128	Save report as: `research_{topic}_{YYYYMMDD}.md`
129
130	### Formatting Rules
131
132	- Chinese content: Use full-width punctuation（，。：；！？）
133	- Technical terms: Provide Wiki/doc URL on first mention
134	- Tables: Use for metrics, comparisons
135	- Code blocks: For technical examples
136	- Mermaid: For architecture, timelines, flows
137
138	## Best Practices
139
140	1. Start with official sources - Repo, docs, company blog
141	2. Verify dates from commits/PRs - More reliable than articles
142	3. Triangulate claims - 2+ independent sources
143	4. Note conflicting info - Don't hide contradictions
144	5. Distinguish fact vs opinion - Label speculation clearly
145	6. CRITICAL: Always include inline citations - Use `[citation:Title](URL)` format immediately after each claim from external sources
146	7. Extract URLs from search results - web_search returns {title, url, snippet} - always use the URL field
147	8. Update as you go - Don't wait until end to synthesize
148
149	### Citation Examples
150
151	Good - With inline citations:
152	```markdown
153	The project gained 10,000 stars within 3 months of launch [citation:GitHub Stats](https://github.com/owner/repo).
154	The architecture uses LangGraph for workflow orchestration [citation:LangGraph Docs](https://langchain.com/langgraph).
155	```
156
157	Bad - Without citations:
158	```markdown
159	The project gained 10,000 stars within 3 months of launch.
160	The architecture uses Lan

People who liked this also viewed

6 picks

Type

github-deep-research

91stars18forks

View on GitHub

How to install github-deep-research?

opsintech/opsintech-platform/github-deep-research

$npx -y skills add OpsinTech/opsintech-platform --skill github-deep-research

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# GitHub Deep Research Skill
2
3	Multi-round research combining GitHub API, web_search, web_fetch to produce comprehensive markdown reports.
4
5	## Research Workflow
6
7	- Round 1: GitHub API
8	- Round 2: Discovery
9	- Round 3: Deep Investigation
10	- Round 4: Deep Dive
11
12	## Core Methodology
13
14	### Query Strategy
15
16	Broad to Narrow: Start with GitHub API, then general queries, refine based on findings.
17
18	```
19	Round 1: GitHub API
20	Round 2: "{topic} overview"
21	Round 3: "{topic} architecture", "{topic} vs alternatives"
22	Round 4: "{topic} issues", "{topic} roadmap", "site:github.com {topic}"
23	```
24
25	Source Prioritization:
26	1. Official docs/repos (highest weight)
27	2. Technical blogs (Medium, Dev.to)
28	3. News articles (verified outlets)
29	4. Community discussions (Reddit, HN)
30	5. Social media (lowest weight, for sentiment)
31
32	### Research Rounds
33
34	Round 1 - GitHub API
35	Directly execute `scripts/github_api.py` without `read_file()`:
36	```bash
37	python /path/to/skill/scripts/github_api.py <owner> <repo> summary
38	python /path/to/skill/scripts/github_api.py <owner> <repo> readme
39	python /path/to/skill/scripts/github_api.py <owner> <repo> tree
40	```
41
42	Available commands (the last argument of `github_api.py`):
43	- summary
44	- info
45	- readme
46	- tree
47	- languages
48	- contributors
49	- commits
50	- issues
51	- prs
52	- releases
53
54	Round 2 - Discovery (3-5 web_search)
55	- Get overview and identify key terms
56	- Find official website/repo
57	- Identify main players/competitors
58
59	Round 3 - Deep Investigation (5-10 web_search + web_fetch)
60	- Technical architecture details
61	- Timeline of key events
62	- Community sentiment
63	- Use web_fetch on valuable URLs for full content
64
65	Round 4 - Deep Dive
66	- Analyze commit history for timeline
67	- Review issues/PRs for feature evolution
68	- Check contributor activity
69
70	## Report Structure
71
72	Follow template in `assets/report_template.md`:
73
74	1. Metadata Block - Date, confidence level, subject
75	2. Executive Summary - 2-3 sentence overview with key metrics
76	3. Chronological Timeline - Phased breakdown with dates
77	4. Key Analysis Sections - Topic-specific deep dives
78	5. Metrics & Comparisons - Tables, growth charts
79	6. Strengths & Weaknesses - Balanced assessment
80	7. Sources - Categorized references
81	8. Confidence Assessment - Claims by confidence level
82	9. Methodology - Research approach used
83
84	### Mermaid Diagrams
85
86	Include diagrams where helpful:
87
88	Timeline (Gantt):
89	```mermaid
90	gantt
91	title Project Timeline
92	dateFormat YYYY-MM-DD
93	section Phase 1
94	Development :2025-01-01, 2025-03-01
95	section Phase 2
96	Launch :2025-03-01, 2025-04-01
97	```
98
99	Architecture (Flowchart):
100	```mermaid
101	flowchart TD
102	A[User] --> B[Coordinator]
103	B --> C[Planner]
104	C --> D[Research Team]
105	D --> E[Reporter]
106	```
107
108	Comparison (Pie/Bar):
109	```mermaid
110	pie title Market Share
111	"Project A" : 45
112	"Project B" : 30
113	"Others" : 25
114	```
115
116	## Confidence Scoring
117
118	Assign confidence based on source quality:
119
120	\| Confidence \| Criteria \|
121	\|------------\|----------\|
122	\| High (90%+) \| Official docs, GitHub data, multiple corroborating sources \|
123	\| Medium (70-89%) \| Single reliable source, recent articles \|
124	\| Low (50-69%) \| Social media, unverified claims, outdated info \|
125
126	## Output
127
128	Save report as: `research_{topic}_{YYYYMMDD}.md`
129
130	### Formatting Rules
131
132	- Chinese content: Use full-width punctuation（，。：；！？）
133	- Technical terms: Provide Wiki/doc URL on first mention
134	- Tables: Use for metrics, comparisons
135	- Code blocks: For technical examples
136	- Mermaid: For architecture, timelines, flows
137
138	## Best Practices
139
140	1. Start with official sources - Repo, docs, company blog
141	2. Verify dates from commits/PRs - More reliable than articles
142	3. Triangulate claims - 2+ independent sources
143	4. Note conflicting info - Don't hide contradictions
144	5. Distinguish fact vs opinion - Label speculation clearly
145	6. CRITICAL: Always include inline citations - Use `[citation:Title](URL)` format immediately after each claim from external sources
146	7. Extract URLs from search results - web_search returns {title, url, snippet} - always use the URL field
147	8. Update as you go - Don't wait until end to synthesize
148
149	### Citation Examples
150
151	Good - With inline citations:
152	```markdown
153	The project gained 10,000 stars within 3 months of launch [citation:GitHub Stats](https://github.com/owner/repo).
154	The architecture uses LangGraph for workflow orchestration [citation:LangGraph Docs](https://langchain.com/langgraph).
155	```
156
157	Bad - Without citations:
158	```markdown
159	The project gained 10,000 stars within 3 months of launch.
160	The architecture uses Lan

People who liked this also viewed

6 picks

Type