Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Execute this skill enhances AI assistant's ability to conduct web research and translate findings into actionable github issues. it automates the process of extracting key information from web search results and formatting it into a well-structured issue, ready... Use when managing version control. Trigger with phrases like 'commit', 'branch', or 'git'.
.claude/skills/jeremylongshore-creating-github-issues-from-web-research/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 458% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -16% | 0% |
| case-04 | ✓→✗ | ▼ Worse | 2% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 114% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | 5% | 0% |
Convert web research findings into well-structured GitHub issues with titles, summaries, key recommendations, source links, and appropriate labels.
Streamline the research-to-implementation workflow. By integrating web search with GitHub issue creation, Claude can efficiently convert research findings into trackable tasks for development teams.
This skill activates when you need to:
User request: "research Docker security best practices and create a ticket in myorg/backend"
The skill will:
User request: "find articles about API rate limiting, create issue with label performance"
The skill will:
This skill seamlessly integrates with Claude's web search Skill and requires authentication with a GitHub account. It can be used in conjunction with other skills to further automate development workflows.
The skill produces structured output relevant to the task.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 16,495 | 15,129 | -8% | 1 | 1 | 0% | 3,062 | 2,879 | -6% | 0 | 0 | — |
case-02 | pass→pass | 3,939 | 4,533 | +15% | 1 | 1 | 0% | 624 | 1,337 | +114% | 0 | 0 | — |
case-03 | pass→pass | 15,206 | 10,786 | -29% | 1 | 1 | 0% | 2,431 | 2,544 | +5% | 0 | 0 | — |
case-04 | pass→fail | 5,266 | 6,333 | +20% | 1 | 1 | 0% | 934 | 949 | +2% | 0 | 0 | — |
case-05 | fail→fail | 6,554 | 10,988 | +68% | 1 | 1 | 0% | 489 | 2,128 | +335% | 0 | 0 | — |
case-06 | fail→fail | 16,186 | 14,241 | -12% | 1 | 1 | 0% | 3,653 | 3,434 | -6% | 0 | 0 | — |
case-07 | fail→pass | 7,664 | 9,070 | +18% | 1 | 1 | 0% | 395 | 2,203 | +458% | 0 | 0 | — |
case-08 | fail→fail | 17,867 | 6,625 | -63% | 1 | 1 | 0% | 3,111 | 1,149 | -63% | 0 | 0 | — |
case-14 | fail→fail | 4,543 | 8,073 | +78% | 1 | 1 | 0% | 249 | 1,167 | +369% | 0 | 0 | — |
case-09 | fail→fail | 13,085 | 16,793 | +28% | 1 | 1 | 0% | 2,448 | 2,740 | +12% | 0 | 0 | — |
case-10 | fail→fail | 16,258 | 11,213 | -31% | 1 | 1 | 0% | 2,779 | 2,710 | -2% | 0 | 0 | — |
case-11 | fail→fail | 14,701 | 9,748 | -34% | 1 | 1 | 0% | 2,592 | 1,311 | -49% | 0 | 0 | — |
case-12 | fail→fail | 4,192 | 11,393 | +172% | 1 | 1 | 0% | 260 | 3,054 | +1075% | 0 | 0 | — |
case-13 | fail→fail | 13,840 | 14,441 | +4% | 1 | 1 | 0% | 2,498 | 3,600 | +44% | 0 | 0 | — |
case-15 | fail→pass | 18,417 | 17,385 | -6% | 1 | 1 | 0% | 3,752 | 3,156 | -16% | 0 | 0 | — |
case-16 | fail→fail | 20,846 | 4,779 | -77% | 1 | 1 | 0% | 2,915 | 801 | -73% | 0 | 0 | — |
case-17 | fail→fail | 17,485 | 13,652 | -22% | 1 | 1 | 0% | 2,977 | 3,051 | +2% | 0 | 0 | — |
case-18 | fail→fail | 9,074 | 11,327 | +25% | 1 | 1 | 0% | 339 | 2,669 | +687% | 0 | 0 | — |
case-19 | fail→fail | 13,853 | 11,570 | -16% | 1 | 1 | 0% | 2,382 | 2,820 | +18% | 0 | 0 | — |
case-20 | fail→fail | 8,377 | 16,074 | +92% | 1 | 1 | 0% | 683 | 2,591 | +279% | 0 | 0 | — |
case-21 | fail→fail | 5,929 | 10,805 | +82% | 1 | 1 | 0% | 412 | 1,322 | +221% | 0 | 0 | — |
case-22 | fail→fail | 28,918 | 6,254 | -78% | 1 | 1 | 0% | 5,314 | 1,054 | -80% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 10 counted toward the lift figure. The other 12 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +5 percentage points is the difference between those two pass rates over the 10 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.