ark-research

Research technical solutions by searching the web, examining GitHub repos, and gathering evidence. Use when exploring implementation options or evaluating technologies.

1.65x

Quality

58%

Does it follow best practices?

Impact

68%

1.65x

Average score across 3 eval scenarios

Securityby

Advisory

Suggest reviewing before use

Optimize this skill with Tessl

npx tessl skill review --optimize ./.claude/skills/research/SKILL.md

Evaluation results

100%

53%

Python Async Job Queue Evaluation

Local research workspace and evidence requirements

Criteria

Without context

With context

Scratch directory created

100%

Files saved in scratch

100%

Multiple scratch files

100%

Minimum 2 datapoints

100%

Minimum 3 datapoints

100%

GitHub repo evidence

62%

100%

Official docs evidence

100%

Real-world usage evidence

37%

100%

Comparison notes saved

100%

Code snippets saved

100%

Recommendation present

100%

62%

28%

WebSocket Pub/Sub Server Research for Node.js

GitHub repo examination and structured output format

Criteria

Without context

With context

Repo cloned to /tmp

100%

README examined

100%

Code/architecture examined

100%

Dependencies examined

100%

Research heading format

Option sections with Source

50%

Option sections with Pros and Cons

25%

Option sections with Evidence

Recommendation section

71%

100%

Source count in recommendation

Sources list with type labels

100%

Web search evidence

100%

42%

-2%

Distributed Tracing for Python Microservices

Full research workflow with web search and output format

Criteria

Without context

With context

Scratch workspace used

Research heading format

Option Source fields

Option Pros fields

100%

71%

Option Cons fields

100%

71%

Option Evidence fields

Recommendation section

57%

71%

Source count in recommendation

Sources section present

57%

71%

Sources with type annotations

Minimum 2 evidence sources

100%

Minimum 3 evidence sources

100%

GitHub repo referenced

100%

Repository: mckinsey/agents-at-scale-ark
Commit: f4bfd2d

Evaluated: 30 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Python Async Job Queue Evaluation WebSocket Pub/Sub Server Research for Node.js Distributed Tracing for Python Microservices

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.