browse-and-evaluate — Claude Code skill: search and install
The browse-and-evaluate skill helps you find, compare, and inspect catalog skills before adding them. Use it when catalog output needs to stay focused.
MIT 177 installs ★ 1,161 updated 2026-09-21
npx skills add MoizIbnYousaf/ai-agent-skills --skill browse-and-evaluate -a claude-codeWhat browse-and-evaluate does
The skill defines a catalog workflow for ai-agent-skills: search with search, inspect candidate details through info, then open the content with preview. It keeps command output narrow with --fields and uses --limit when a catalog view is large.
Before any install, it runs npx ai-agent-skills install <skill-name> --dry-run; the real install follows only after that result has been reviewed. Collections should be checked with --list or --dry-run, while upstream skills need a network fetch during installation. Skipping the preliminary --dry-run breaks the sequence enforced by the skill.
It can be used with Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, and OpenCode. For non-interactive pipelines, the skill recommends JSON output.
What it helps with
- search the catalog for a task-specific skill
- compare several catalog candidates
- inspect skill content before adding it
- preview an install with
--dry-run - keep CLI results within a set limit
Example prompts for the agent
- Find catalog skills for PDF work and return only the name, tier, work area, and description.
- Compare candidates for browser testing, then show the details and preview for each one.
- Run a
--dry-runcheck forsome-skilland ask before adding more than 3 skills.
How to install browse-and-evaluate
1Skills CLI (Vercel)
npx skills add MoizIbnYousaf/ai-agent-skills --skill browse-and-evaluate -a claude-code2Manually: put the files in place
~/.claude/skills/browse-and-evaluate/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/browse-and-evaluate/
1Skills CLI (Vercel)
npx skills add MoizIbnYousaf/ai-agent-skills --skill browse-and-evaluate -a codex2Manually: put the files in place
.agents/skills/browse-and-evaluate/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add MoizIbnYousaf/ai-agent-skills --skill browse-and-evaluate -a cursor2Manually: put the files in place
.cursor/skills/browse-and-evaluate/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install MoizIbnYousaf/ai-agent-skills skills/browse-and-evaluate2Manually: put the files in place
.github/skills/browse-and-evaluate/SKILL.md1Skills CLI (Vercel)
npx skills add MoizIbnYousaf/ai-agent-skills --skill browse-and-evaluate -a gemini-cli2Platform CLI
gemini skills install https://github.com/MoizIbnYousaf/ai-agent-skills --path skills/browse-and-evaluateManually: put the files in place
.devin/skills/browse-and-evaluate/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add MoizIbnYousaf/ai-agent-skills --skill browse-and-evaluate -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about browse-and-evaluate
How do I install browse-and-evaluate in Claude Code?
Run npx skills add MoizIbnYousaf/ai-agent-skills --skill browse-and-evaluate -a claude-code. The skill file will be placed at ~/.claude/skills/browse-and-evaluate/SKILL.md.
What does browse-and-evaluate do?
It guides catalog search, candidate inspection, content preview, and install checks before a skill is added.
Why does it use --fields and --dry-run?
--fields keeps search, info, and list output small. --dry-run checks the install source and planned changes before the actual command runs.
Source text
SKILL.md · MIT
--- name: browse-and-evaluate description: Use when exploring the ai-agent-skills catalog to find, compare, and evaluate skills before installing. Always use --fields to limit output size and --dry-run before committing to an install. category: workflow version: 4.1.0 --- # Browse And Evaluate ## Goal Find the right skill for a task without flooding the context window or installing blindly. ## Guardrails - Always use `--fields` on list/search/info to keep output small. Default: `--fields name,tier,workArea,description`. - Always use `--dry-run` before installing anything. - Never install more than 3 skills at once without explicit user confirmation. - Prefer `--format json` in non-interactive pipelines. The CLI defaults to JSON when stdout is not a TTY. - Use `--limit` when browsing large catalogs. Start with `--limit 10`. ## Workflow 1. Search or browse the catalog. ```bash npx ai-agent-skills search <query> --fields name,tier,workArea,description --limit 10 ``` 2. Get details on a candidate. ```bash npx ai-agent-skills info <skill-name> --fields name,description,tags,collections,installCommands ``` 3. Preview the skill content. ```bash npx ai-agent-skills preview <skill-name> ``` 4. Dry-run the install. ```bash npx ai-agent-skills install <skill-name> --dry-run ``` 5. Install only after reviewing the dry-run output. ```bash npx ai-agent-skills install <skill-name> ``` ## Gotchas - The `preview` command sanitizes skill content to strip prompt injection patterns. If content looks truncated, check if suspicious patterns were removed. - Collection installs pull multiple skills. Always `--list` or `--dry-run` a collection before installing. - Upstream (non-vendored) skills require a network fetch at install time. Use `--dry-run` to verify the source is reachable.
© MoizIbnYousaf · MIT · MoizIbnYousaf/ai-agent-skills