monkey1wizard

monkey1wizard/Golem-Agents-Legion

Agent Claude Code

Business Analyst that reviews plans and features from a commercial perspective — validates business logic, ROI, user impact, and market fit. Provides business-oriented recommendations.

monkey1wizard/Golem-Agents-Legion

Agent Claude Code

Strict, neutral Technical Architect that reviews plans and ideas for trade-offs, over-engineering, bugs, and performance. Adversarial counterpart to the planning author — pushes back on bad decisions.

monkey1wizard/Golem-Agents-Legion

Agent Claude Code

Performs per-task audit for deep performance issues and security findings, then writes only high-confidence findings back to the active plan.

monkey1wizard/Golem-Agents-Legion

Agent Claude Code

Investigates bugs with scientific method, internal freeze discipline, and root-cause confirmation before any fix.

monkey1wizard/Golem-Agents-Legion

Agent Claude Code

Owns GAL's design system, variant exploration, design-to-code build, and live UI audit for customer-facing work.

monkey1wizard/Golem-Agents-Legion

Agent Claude Code

Executes approved plan files with atomic commits, deviation handling, and plan-level state tracking. Updates plan.

monkey1wizard/Golem-Agents-Legion

Agent Claude Code

Planning-stage release-flow designer (consult). Researches API/MCP/CICD tools and designs a release/devops flow for the project; emits design advice. Does not write plan files and does not execute.

monkey1wizard/Golem-Agents-Legion

Agent Claude Code

Dedicated research investigator for RESEARCH, SYNTHESIZE, and CROSS-REVIEW states. Uses local-first search, source attribution, and structured synthesis before independent reference verification and documentation.

monkey1wizard/Golem-Agents-Legion

Agent Claude Code

First-class documentation-structure steward. Owns the NDJSON structure map, code→doc drift detection, end-of-run knowledge extraction into docs/, docs/ and .dev/plans/ structural hygiene, and figure/flowchart sync. Callable via /gal steward.

monkey1wizard/Golem-Agents-Legion

Agent Claude Code

Owns spec-driven verification and real-browser QA. Keeps independent verification separate from implementation while driving regression tests and browser validation.

monkey1wizard/Golem-Agents-Legion

Command Claude Code

GAL now exposes a deliberately small public command surface.

monkey1wizard/Golem-Agents-Legion

Skill Claude Code

Single-source adversarial review method for plans, diffs, docs, decisions, and claims: steel-man first, refute by default under doubt, enforce evidence discipline, use APPROVE/REVISE/REJECT verdicts, stop at jidoka boundaries, and never treat NotRun as pass.

monkey1wizard/Golem-Agents-Legion

Skill Claude CodeCodex

Extract clean markdown content from web pages using Defuddle CLI, removing clutter and navigation to save tokens. Use instead of WebFetch when the user provides a URL to read or analyze, for online documentation, articles, blog posts, or any standard web page.

monkey1wizard/Golem-Agents-Legion

Skill Claude CodeCodex

Guide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers.…

monkey1wizard/Golem-Agents-Legion

Skill Claude CodeCodex

Maintain the dual-axis NDJSON structure map for docs drift detection and sync. Use when code changes require doc-impact mapping, per-line NDJSON updates, reconcile scans, or live projection rendering from structure-map.ndjson.

monkey1wizard/Golem-Agents-Legion

Skill Claude Code

Write commit messages that follow this project's Conventional Commit format, and when the user's intent is to actually commit staged changes, generate the message via git-commit-msg rules and execute git commit with that exact message. Use whenever the user asks for a commit message, says git-commit-msg, git-commit…

monkey1wizard/Golem-Agents-Legion

Skill Claude CodeCodex

An installation procedure for the GAL command-line program and its machine setup. It checks whether GAL is already available, chooses an installation path for the operating system, and verifies the result.

monkey1wizard/Golem-Agents-Legion

Skill Claude CodeCodex

Search local repo-owned material first, then use optional local-notes backends only when they are ready. Activate before answering conceptual questions, proposing solutions, or planning note-adjacent work.

monkey1wizard/Golem-Agents-Legion

Skill Claude CodeCodex

Format Markdown consistently for this project. Use when writing or editing Markdown files, documentation, plans, notes, or README content.

monkey1wizard/Golem-Agents-Legion

Skill Claude CodeCodex

Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).

monkey1wizard/Golem-Agents-Legion

Skill Claude CodeCodex

Use OpenCLI for low-token, structured external retrieval when an existing site adapter matches the task. Prefer for repeatable research queries such as YouTube transcript, NotebookLM source access, Wikipedia summary, Hacker News search, and similar structured data retrieval tasks.

monkey1wizard/Golem-Agents-Legion

Skill Claude CodeCodex

Apply the Result pattern for expected failures and reserve exceptions for unexpected faults. Use when designing error handling, validation flows, parsing, or business-rule outcomes.

monkey1wizard/Golem-Agents-Legion

Skill Claude CodeCodex

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

monkey1wizard/Golem-Agents-Legion

Skill Claude CodeCodex

Apply structured logging conventions across languages. Use when adding logs, reviewing logging quality, or choosing log levels and log field structure.