Superclaw

Structured development workflow for OpenClaw agents - Brainstorm, Plan, Execute, Review. Brings systematic software development with TDD, clear specs, and su...

Security Scan

This is an instruction-only workflow skill that coherently documents a TDD-based agent development process and only requires git; nothing requested or described is disproportionate to that purpose.

Purpose & Capability

Name and description match the actual instructions: a four-phase TDD development workflow that creates design/plan docs, runs tasks, and makes git commits. The only required binary is git, which is appropriate.

Instruction Scope

SKILL.md provides high-level, expected workflow steps (create files under docs/, write tests, make commits, review). It does not instruct reading unrelated files or accessing secrets. One vague item — 'reports progress' — could be interpreted broadly (e.g., sending progress to an external endpoint) and should be clarified before granting autonomous network permissions.

Install Mechanism

No install spec or code is present (instruction-only), so nothing will be downloaded or written by an installer. This is the lowest-risk install model.

Credentials

No environment variables, credentials, or config paths are required. Requesting only git is proportionate to the stated purpose.

Persistence & Privilege

always is false and the skill is user-invocable. It does not request persistent elevated privileges or modifications to other skills or system-wide settings.

Assessment

This skill appears coherent and low-risk, but before installing: (1) confirm what 'reports progress' means and whether the agent will be allowed to send data externally; (2) verify whether the agent is permitted to push commits to remote repos — if you don't want remote pushes, ensure network or git credentials are restricted; (3) ensure your repo/workspace is backed up and that any automated tests/tools the workflow expects are available; (4) run the skill in an isolated environment or with limited network access if you want to be cautious. Additionally, note the README references an MIT license but no LICENSE file is included in the package — if license matters, request the license file from the author.

SuperClaw - Agent Development Workflow

SuperClaw brings systematic, professional software development to OpenClaw agents. Based on proven patterns from Superpowers, adapted for the agent community.

Philosophy

• Test-Driven Development - Write tests first, always
• Systematic over ad-hoc - Process over guessing
• Complexity reduction - Simplicity as primary goal
• Evidence over claims - Verify before declaring success

Workflow Overview

SuperClaw follows a 4-phase workflow:

Brainstorm → Plan → Execute → Review

Phase 1: Brainstorm 🧠

When to use: Before writing any code

What it does:

• Asks clarifying questions about requirements
• Explores alternative approaches
• Identifies risks and constraints
• Creates a design document in chunks

Output: docs/superclaw/designs/YYYY-MM-DD-<feature>.md

Key Principle: Don't jump into code. Understand the problem first.

Phase 2: Write Plan 📝

When to use: After design approval

What it does:

• Breaks work into bite-sized tasks (2-5 minutes each)
• Maps file structure and responsibilities
• Specifies exact test criteria
• Includes verification steps

Output: docs/superclaw/plans/YYYY-MM-DD-<feature>.md

Phase 3: Execute ⚡

When to use: After plan approval

What it does:

• Works through tasks sequentially
• Enforces RED-GREEN-REFACTOR TDD cycle
• Makes frequent commits
• Reports progress

Phase 4: Review ✅

When to use: Between tasks and at completion

What it does:

• Reviews code against plan
• Checks for spec compliance
• Identifies code quality issues
• Blocks progress on critical issues

---

Credits

Adapted from Superpowers by Jesse Vincent (obra). Reimagined for the OpenClaw and ChatClaw agent community.

MIT License - see LICENSE file