browser-probe

Exploratory browser QA for real web apps.

IndexedMIT

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/mylesmcook-browser-probe

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

About

Exploratory browser QA for real web apps. Use when you want a smoke test, scoped verification, exploratory testing, or a balanced read on what works, what feels off, and what is blocked. Prefer this over `dogfood` when the user wants a lighter pass instead of handoff-grade repro artifacts.

Use cases: testing

Reviews

Write a review

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools: No tools found
Secrets: None asked for
Install scripts: None found

Network hosts it talks to 1

codeload.github.com · found in install command

Files it touches

~/.claude/skills/browser-probe

What's inside

Skill name: browser-probe
Description: Exploratory browser QA for real web apps. Use when you want a smoke test, scoped verification, exploratory testing, or a balanced read on what works, what feels off, and what is blocked. Prefer this over `dogfood` when the user wants a lighter pass instead of handoff-grade repro artifacts.
License: MIT
Instructions: 35 lines in SKILL.md · view on GitHub

Files (14)

SKILL.md
metadata.json
auth.md · references/patterns/
data-display.md · references/patterns/
forms.md · references/patterns/
navigation.md · references/patterns/
overlays.md · references/patterns/
search-filter.md · references/patterns/
recovery.md · references/
session-preflight.md · references/
stage-1-model.md · references/
stage-2-hypothesize.md · references/
stage-3-execute.md · references/
stage-4-report.md · references/

SKILL.md (first lines)

# Browser Probe

Use this skill for exploratory browser QA that stays lighter than a formal repro package.

## Workflow

1. Pick a mode: `smoke`, `balanced`, or `bug-hunt`.
2. Create one named `agent-browser` session and reuse it for the whole run.
3. Model the app or scoped feature.
4. Generate likely failure hypotheses and likely strengths.
5. Execute focused checks with proportional evidence.
6. Report confirmed issues, verified strengths, flakes, and blocked areas.
7. Escalate to `dogfood` if the user now needs handoff-grade repro artifacts.

## References

Read these in order:
- `references/stage-1-model.md`
- `references/stage-2-hypothesize.md`
- `references/stage-3-execute.md`
- `references/stage-4-report.md`

Use these only when needed:
- `references/session-preflight.md`
- `references/recovery.md`
- `references/patterns/*`

## Guardrails

- Keep one live browser session per run.
- Re-snapshot after DOM-changing actions.
- Use proportional evidence, not artifact theater.
- Route repro-heavy handoff requests to `dogfood`.

From MylesMCook/mcook-skills, MIT.

Signals

Stars: 0
Forks: 0
Open issues: 0
Last commit (this folder): Apr 22, 2026 (6 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License: MIT
Repository: github.com/MylesMCook/mcook-skills
First seen: Oct 24, 2026
Last checked: Oct 24, 2026
Source: GitHub github-graphql

More from MylesMCook

Cite this page

browser-probe skill. Agent Mart. https://agentmart.dev/skills/mylesmcook-browser-probe (last updated 2026-10-24)