SuperAPI AI Directory

technical-research

Writing & research

View repository

Updated 10-15 22:15

About this skill

Technical Research

Owns evidence-based investigation: verifying claims, comparing options with sourced tradeoffs, and confirming how libraries, APIs, and tools actually behave before conclusions are relied on.

1. Discipline

  • Verify information from multiple sources before presenting as fact
  • Prefer official documentation over blog posts; note publication dates
  • Flag confidence level (High/Medium/Low) on each finding
  • Clearly separate facts from opinions and recommendations

2. Workflow

  1. Clarify scope: what exactly needs to be answered?
  2. Search official docs first (WebSearch + WebFetch)
  3. Check GitHub repos: issues, discussions, source code for implementation details
# Clone to /tmp for analysis when needed
git clone --depth 1 <repo-url> /tmp/<repo-name>
  1. Cross-reference: do multiple sources agree? Flag contradictions.
  2. Test when possible: run code, check versions, verify claims hands-on
  3. When synthesis needs categories, labels, or first-use terms before final wording, use clarify-concepts after evidence acquisition and before drafting conclusions.

3. Source Priority

  1. Official documentation and changelogs
  2. Source code and GitHub issues/discussions
  3. Well-maintained community resources
  4. Blog posts (note date, may be outdated)

4. Output Format