Skip to content

nsudhanva/sudhanva-agent-skills

v1.0.0MIT

Read-only skills and MCP connections for researching Sudhanva Narayana's public profile, production ML work, and technical writing.

GitHubOpen Website

Sudhanva Agent Skills

Three small, read-only Agent Skills for researching Sudhanva Narayana's public work. Each skill uses canonical sudhanva.me pages and APIs and includes explicit guidance against inventing unpublished details.

Install

Install all three skills:

npx skills add nsudhanva/sudhanva-agent-skills --all

Or select an individual skill when prompted:

npx skills add nsudhanva/sudhanva-agent-skills

Included skills

SkillPurpose
research-sudhanva-profileResearch the public professional profile and experience.
find-production-ml-case-studiesFind and compare published production ML case studies.
find-technical-writingSearch and cite published technical writing.

Source and safety

  • All referenced data is public and read-only.
  • Prefer canonical pages and API responses over search snippets.
  • Preserve qualifiers on quantitative outcomes.
  • Do not infer private contact details, availability, compensation, client names, or unpublished infrastructure.
  • The canonical discovery index is /.well-known/agent-skills/index.json.
  • plugin.json and mcp.json bundle these skills with the public-data and profile-insight MCP servers as an Agent Plugin.

License

MIT. See LICENSE.

Validation

Run the same dependency-free checks used by CI:

npm test

The suite verifies the plugin and MCP manifests, canonical project URLs, and every skill's frontmatter, directory name, top-level heading, and README entry.