github-deep-research

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.

Quality

75%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/public/github-deep-research/SKILL.md

Quality

Content

67%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, largely actionable research workflow with a clear four-round sequence and real, correctly documented bundle files. The main drag is token efficiency: generic Mermaid syntax tutorials and a report outline that duplicates assets/report_template.md inflate the body without adding knowledge Claude lacks.

Suggestions

Replace the three full Mermaid example blocks with one-line guidance ('use gantt for timelines, flowchart for architecture, pie for comparisons') since Claude already knows Mermaid syntax.

Collapse the 9-item Report Structure section to a single pointer: 'Follow the full template in assets/report_template.md' - the template already enumerates every section.

Promote verification from Best Practices into the workflow itself, e.g. after Round 3: 'Cross-check each timeline date against commit/release data before writing the report.'

Dimension	Reasoning	Score
Conciseness	Mostly efficient, but roughly 30 lines of generic Mermaid syntax examples (gantt/flowchart/pie) explain syntax Claude already knows, and the 9-item 'Report Structure' list duplicates assets/report_template.md. Not 4: the Mermaid section plus the duplicated report outline is more than minor over-explanation; not 2: the rounds, commands, citation rules, and confidence table are all lean and load-bearing.	3 / 5
Actionability	Provides copy-paste-ready commands ('python /path/to/skill/scripts/github_api.py <owner> <repo> summary') that match the real script's CLI, plus a command list, query templates, citation format, and output naming convention. Not 5: Rounds 2-4 are high-level bullet guidance ('Get overview and identify key terms') rather than executable specifics.	4 / 5
Workflow Clarity	A clear four-round sequence with per-round search budgets ('3-5 web_search', '5-10 web_search + web_fetch') and verification guidance via 'Triangulate claims - 2+ independent sources' and confidence scoring. Not 5: verification lives in Best Practices rather than as explicit checkpoints wired into the round transitions; not 3: the sequence is explicit and checks are present.	4 / 5
Progressive Disclosure	Good structure with both bundle references real, one level deep, and clearly signaled (scripts/github_api.py commands verified against the actual script; assets/report_template.md verified to exist). Not 5: the Mermaid examples and the Report Structure list duplicate template content that could live in the referenced file; not 3: no large inline content that clearly belongs in a separate file and no buried references.	4 / 5
	Total	15 / 20 Passed

Description

83%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description that clearly states what the skill does, what it produces, and when to use it, with concrete trigger phrases. Main weaknesses are missing synonym-level trigger coverage and minor overlap risk from broad phrases like 'open source projects'.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete deliverables ('executive summaries, chronological timelines, metrics analysis, and Mermaid diagrams'), matching the 'several specific actions; minor gaps' anchor. It falls short of 5 because the underlying research actions (commit/issue/PR analysis) are not enumerated, and 'conduct deep research' remains somewhat abstract.	4 / 5
Completeness	Explicitly answers both 'what' ('Conduct multi-round deep research... Produces structured markdown reports with...') and 'when' ('Use when users request comprehensive analysis... Triggers on Github repository URL or open source projects') with concrete trigger phrases. Not 4: the 'when' clause is already explicit and specific, not merely implied.	5 / 5
Trigger Term Quality	Good keyword coverage with natural phrases ('comprehensive analysis', 'timeline reconstruction', 'competitive analysis', 'GitHub repository URL', 'open source projects'). Not 5: common variations like 'analyze this repo' or 'repository research' are missing and the GitHub casing is inconsistent.	4 / 5
Distinctiveness Conflict Risk	Mostly distinct niche (GitHub repo deep research) with a highly specific URL trigger, but broad phrases like 'comprehensive analysis' and 'open source projects' create minor overlap risk with a generic deep-research skill. Not 5 due to that overlap; not 3 since the GitHub-repository-URL trigger is unlikely to fire for unrelated skills.	4 / 5
	Total	17 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 1 missing	Warning

	Total	15 / 16 Passed

Repository: bytedance/deer-flow
Path: skills/public/github-deep-research/SKILL.md
Commit: ee44d1e

Reviewed: 18 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.