Hands On Test

What it does

When your agent uses it

• The user wants to visually verify a web feature
• Perform exploratory testing
• Validate UI behavior

Example prompts

• “Use the hands-on-test skill to perform manual hands-on testing of a web application using playwright-cli”
• “/hands-on-test”

Requirements

• Node.js
• Docker
• Pre-approved tools (allowed-tools): Bash(playwright-cli:*), Bash(docker compose:*), Bash(lsof:*), Bash(curl:*), Bash(mkdir:*), Bash(date:*), Read, Write

Workflow steps

5 steps, taken from the step headings in SKILL.md.

1. 1Environment Setup
2. 2Browser Session
3. 3Outcome Verification
4. 4Cleanup
5. 5Report Generation

What it can do on your machine

Read from SKILL.md and the folder at commit 93414a1. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves these tools, so the agent can use them without asking each time:

  • Bash(playwright-cli:*)
  • Bash(docker compose:*)
  • Bash(lsof:*)
  • Bash(curl:*)
  • Bash(mkdir:*)
  • Bash(date:*)
  • Read
  • Write

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • curl
  • docker

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md. Its commands use curl and docker, which can reach the network depending on how they are called.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Hands On Test loads about 1.7k tokens when it runs. Until then it costs about 90 tokens; SKILL.md has 351 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~90

When it runs · the whole SKILL.md, loaded when a task matches

~1.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from ktnyt/cclsp at commit 93414a1, republished under its MIT licence (© ktnyt). 351 words, ~1,690 tokens.

name

hands-on-test

description

Performs manual hands-on testing of a web application using playwright-cli. Spawns the dev server if needed, navigates to pages, performs browser actions, captures screenshots, checks outcomes, and produces a structured test report. Use when the user wants to visually verify a web feature, perform exploratory testing, or validate UI behavior.

allowed-tools

Bash(playwright-cli:*), Bash(docker compose:*), Bash(lsof:*), Bash(curl:*), Bash(mkdir:*), Bash(date:*), Read, Write

TEST_OUTPUT_DIR="$CLAUDE_PROJECT_DIR/test-output/$(date +%Y%m%d%H%M)"
mkdir -p "$TEST_OUTPUT_DIR"

# Check if something is listening on the expected port
lsof -i :3000 -sTCP:LISTEN
# or
curl -s -o /dev/null -w "%{http_code}" http://localhost:3000

# Example: Docker Compose project
docker compose up -d

# Example: Node.js project
# npm run dev &

# Poll until the server responds
for i in $(seq 1 30); do
  curl -s -o /dev/null -w "%{http_code}" http://localhost:3000 | grep -q "200" && break
  sleep 1
done

playwright-cli open http://localhost:3000

playwright-cli open http://localhost:3000/path/to/test

playwright-cli snapshot

# Click elements
playwright-cli click e3

# Fill form fields
playwright-cli fill e5 "test input"

# Type text (simulates keystrokes)
playwright-cli type "search query"

# Press keys
playwright-cli press Enter

# Select dropdown options
playwright-cli select e9 "option-value"

# Check/uncheck checkboxes
playwright-cli check e12
playwright-cli uncheck e12

# Hover
playwright-cli hover e4

# Navigate
playwright-cli goto http://localhost:3000/other-page

playwright-cli snapshot

# Full page screenshot
playwright-cli screenshot --filename=$TEST_OUTPUT_DIR/<test-name>-<step>.png

# Screenshot of a specific element
playwright-cli screenshot e5 --filename=$TEST_OUTPUT_DIR/<test-name>-<element>.png

playwright-cli console

playwright-cli network

playwright-cli close

# Hands-On Test Report

**Date**: YYYY-MM-DD HH:MM
**Tester**: <agent-name>
**Target**: <URL tested>

## Test Scenario

<Brief description of what was being tested>

## Steps Performed

| # | Action | Target | Details |
|---|--------|--------|---------|
| 1 | Navigate | http://localhost:3000/page | Initial page load |
| 2 | Fill | e5 (Email field) | Entered "user@example.com" |
| 3 | Click | e8 (Submit button) | Submitted form |

## Screenshots

| Step | Screenshot | Description |
|------|-----------|-------------|
| 1 | ![Step 1](test-name-01-initial.png) | Page after initial load |
| 2 | ![Step 2](test-name-02-filled.png) | Form with data entered |
| 3 | ![Step 3](test-name-03-submitted.png) | Result after submission |

## Console Output

## Result

**Status**: PASS / FAIL

<Explanation of whether the outcome matched expectations.
If FAIL, describe the discrepancy between expected and actual behavior.>

# Phase 1: Setup
TEST_OUTPUT_DIR="$CLAUDE_PROJECT_DIR/test-output/$(date +%Y%m%d%H%M)"
mkdir -p "$TEST_OUTPUT_DIR"
curl -s -o /dev/null -w "%{http_code}" http://localhost:3000
# If not running:
docker compose up -d
# Wait for ready
for i in $(seq 1 30); do
  curl -s -o /dev/null -w "%{http_code}" http://localhost:3000 | grep -q "200" && break
  sleep 1
done

# Phase 2: Test
playwright-cli open http://localhost:3000/login
playwright-cli snapshot
playwright-cli screenshot --filename=$TEST_OUTPUT_DIR/login-01-initial.png

playwright-cli fill e5 "user@example.com"
playwright-cli fill e8 "password123"
playwright-cli screenshot --filename=$TEST_OUTPUT_DIR/login-02-filled.png

playwright-cli click e10
playwright-cli snapshot
playwright-cli screenshot --filename=$TEST_OUTPUT_DIR/login-03-result.png

playwright-cli console

# Phase 3: Verify
# Read screenshots to visually verify
# Check snapshot for expected DOM state
# Review console for errors

# Phase 4: Cleanup
playwright-cli close

# Phase 5: Write report to $TEST_OUTPUT_DIR/REPORT.md

© ktnyt, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .claude/skills/hands-on-test of ktnyt/cclsp.

Open the folder on GitHubat commit 93414a1