skills.rest
Unsupervisedcom
UnsupervisedcomOFFICIAL
Explore all skills by this author →
WebsiteGitHub

manual_tests

Automates DeepWork hook and rule tests across isolated sub-agents with enforced constraints.

49|3|Updated Jan 11, 2026
Search Tags:#workflow#hooks#rules#sub-agents#DeepWork#manual-tests
One-click install
npx skills add https://github.com/Unsupervisedcom/deepwork --skill manual-tests
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: manual_tests
Source: https://github.com/Unsupervisedcom/deepwork/tree/main/.claude/skills/manual_tests
Command: npx skills add https://github.com/Unsupervisedcom/deepwork --skill manual-tests

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill simplifies and automates the validation of DeepWork rule systems by orchestrating manual test workflows across sub-agents, ensuring hooks and rules trigger as expected in isolated environments.

Core Features & Use Cases

  • Orchestrated manual testing: Run reset, NOT-fire tests, fire tests, and infinite block tests across sub-agents to validate rule firing behavior.
  • Isolated execution: Sub-agents run in controlled contexts to detect file changes and observe hook outcomes.
  • Configurable reliability: Enforces model and turn limits to prevent hangs and minimize costs.

Quick Start

Use the Skill tool to start with manual_tests.reset, then follow prompts to execute each step of the workflow.

Frequently Asked Questions about manual_tests

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate manual hook and rule tests for DeepWork?▼

DeepWork rule validation is automated by orchestrating sub-agents to run reset, not-fire, fire, and infinite block tests, ensuring hooks trigger correctly and safely in isolated environments without manual file editing.

What types of tests are needed to validate DeepWork rule firing behavior?▼

Validating DeepWork rule firing behavior requires running reset, not-fire, fire, and infinite block tests across sub-agents to verify that hooks fire when expected and remain dormant when they should not trigger.

Do I need to configure sub-agents to run DeepWork manual tests?▼

Yes, running DeepWork manual tests requires enforcing sub-agent constraints by setting the model to haiku and limiting max_turns to 5, which prevents hangs and minimizes costs during isolated test execution.

Why should I use the Skill tool instead of editing files directly for rule testing?▼

Using the Skill tool to trigger test steps is required because it ensures proper orchestration of sub-agents and isolated execution for detecting file changes and observing hook outcomes accurately, rather than editing files directly.

Can I test infinite block scenarios without causing sub-agents to hang?▼

Yes, infinite block tests are executed safely by enforcing strict sub-agent constraints, specifically limiting max_turns to 5 and using the haiku model, which prevents hangs while validating rule behavior.

Related Skills

Explore More Software Engineering Skills→
ncrmroncrmro

deepwork_rules.define

Generate .deepwork/rules/*.md files with YAML frontmatter for file-change triggers.

Community
Advanced
millioncomillionco
14.3k

benchmark-fp-fn-audit

Audit React Doctor diagnostic accuracy against benchmark corpora for false positives and negatives.

Official
Advanced
memorysavermemorysaver

aep-workflow

Construct and manage dynamic multi-agent workflows in Claude Code.

Community
Advanced
UnsupervisedcomUnsupervisedcom
49

manual_tests.run_not_fire_tests

Run six manual safety tests in parallel Haiku sub-agents to verify no rules fire.

Official
Basic