Skip to content

docs: define agent-first CLI support criteria - #558

Open
kaluli123123 wants to merge 1 commit into
addyosmani:mainfrom
kaluli123123:docs/245-agent-first-cli-criteria
Open

kaluli123123 wants to merge 1 commit into
addyosmani:mainfrom
kaluli123123:docs/245-agent-first-cli-criteria

Conversation

@kaluli123123

Copy link
Copy Markdown

Summary

  • add a shared decision reference for recommending agent-safe CLIs
  • document non-interactive, parseable, deterministic, safe-probe, and maintenance requirements
  • give concrete patterns for search, Git, GitHub, static-analysis, and CI commands
  • link the reference from the four skills identified in the issue

Verification

  • node scripts/validate-skills.js (25 skills, 0 errors, 0 warnings)
  • node scripts/run-evals.js (140 checks, 0 errors, rank-1 86%)
  • node scripts/validate-reference-links.js (25 skills, 0 errors)
  • git diff --check
  • all five upstream documentation links returned HTTP 200 on 2026-09-05

Behavioral model grading was not run; descriptions and routing behavior are unchanged.

Closes #245

@addyosmani addyosmani left a comment

Copy link
Copy Markdown
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nice, and it lines up with the build-the-lever idea: criteria for when a skill recommends a CLI (machine-readable output, non-interactive, descriptive errors). It touches a few skills plus a new reference, so it's the most surface in this wave, and I verified it validates. Approving; holding the merge just so it doesn't ride in unnoticed with the smaller docs, happy to land it right after.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Investigate "agent-first" CLIs and document their use across skills

2 participants