Sandgarden AI Software Factory
This repository also configures its own agents. See what sgai tells them →
Skill Claude CodeCodex
Create and setup a new OpenAI Agents SDK application with interactive guidance for language choice, agent type selection (Basic, Voice, Realtime), project setup, and automatic verification.
Skill Claude CodeCodex
Create and setup a new Claude Agent SDK application with interactive guidance for language choice, project setup, and automatic verification.
Skill Claude CodeCodex
Guide for creating and implementing plugins to extend OpenCode functionality, including hooks, custom tools, and event handling. When you need to add custom functionality to OpenCode, such as notifications, custom tools, or modifying behavior.
Skill Claude CodeCodex
Force unrelated concepts together to discover emergent properties - "What if we treated X like Y?". Can't find approach that fits your problem. Conventional solutions feel inadequate. Need innovative solution. Stuck thinking inside one domain. Want breakthrough, not incremental improvement.
Skill Claude CodeCodex
Flip core assumptions to reveal hidden constraints and alternative approaches - "what if the opposite were true?". Stuck on assumptions you can't question. Solution feels forced. "This is how it must be done" thinking. Want to challenge conventional wisdom. Need fresh perspective on problem.
Skill Claude CodeCodex
Spot patterns appearing in 3+ domains to find universal principles. Same issue in different parts of codebase. Pattern feels familiar across projects. "Haven't I solved this before?" Different teams solving similar problems. Recurring solution shapes.
Skill Claude CodeCodex
Test at extremes (1000x bigger/smaller, instant/year-long) to expose fundamental truths hidden at normal scales. Unsure if approach will scale. Edge cases unclear. Want to validate architecture. "Will this work at production scale?" Need to find fundamental limits.
Skill Claude CodeCodex
Find one insight that eliminates multiple components - "if this is true, we don't need X, Y, or Z". Code has many similar-looking implementations. Growing list of special cases. Same concept handled 5 different ways. Excessive configuration. Many if/else branches doing similar things. Complexity spiraling.
Skill Claude CodeCodex
Problem-Solving Dispatch; Dispatch to the right problem-solving technique based on how you're stuck; Stuck on a problem. Conventional approaches not working. Need to pick the right problem-solving technique. Not sure which skill applies.
Skill Claude CodeCodex
Black Swan Analysis provides a rigorous methodology for identifying high-impact, unpredictable events while building systems that thrive under volatility. You must use this framework when the human partner challenges you on your claims, or when the human partner asks you to create plans from vague instructions.
Skill Claude CodeCodex
Interactive idea refinement using Socratic method to develop fully-formed designs. When your human partner says "I've got an idea", "Let's make/build/create", "I want to implement/add", "What if we". When starting design for complex feature. Before writing implementation plans. When idea needs refinement and…
Skill Claude CodeCodex
Interactive wizard to compose valid GOAL.md files for SGAI with step-by-step guidance through 7 phases. Use when creating GOAL.md, configuring agents and model, safety-analysis guidance, or starting an SGAI project. Safety Analysis must use coordinator/reviewer stpa-overview skill guidance.
Skill Claude CodeCodex
Automatically scans GOAL.md for unchecked items, provides completion status summary, and enables coordinator to mark items as complete. When coordinator needs to verify project completion status or before marking work as complete. When coordinator needs to mark completed items in GOAL.md. Symptoms - manually going…
Skill Claude CodeCodex
Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation.
Skill Claude CodeCodex
Use when completing tasks, implementing major features, or before merging to verify work meets requirements - dispatches bounded reviewer subagents with TASKPACKET and REVIEWSTATUS envelopes before proceeding.
Skill Claude CodeCodex
Post-completion factory improvement analysis. Guides the retrospective agent through artifact discovery, pattern analysis, suggestion generation, and presenting proposed changes for approval. Use when the retrospective agent starts its post-completion phase.
Skill Claude CodeCodex
Use when errors occur deep in execution and you need to trace back to find the original trigger - systematically traces bugs backward through call stack, adding instrumentation when needed, to identify source of invalid data or incorrect behavior.
Skill Claude CodeCodex
Guide for using tmux to manage detached sessions for long-running processes, including lifecycle management, cleanup, startup, verification, and output capture; When you need to run servers or long-running commands in the background, send commands to them, capture output, handle TTY requirements, or manage session…
Skill Claude CodeCodex
Coordinator and retrospective workflow-state protocol. Use only when the current agent has permission to call sgaiupdateworkflowstate; Task subagents should return STATUS or REVIEWSTATUS envelopes instead.
Skill Claude CodeCodex
Entry point for STPA (System Theoretic Process Analysis) hazard and safety analysis. Use for full 4-step STPA sessions, focused project-critic safety reviews, coordinator safety gates, reviewer hazard checks, risk assessment, unsafe state transitions, external input, filesystem, concurrency, physical systems, or…
Skill Claude CodeCodex
STPA Step 1 - Define the purpose of analysis by identifying losses, hazards, and system-level constraints. When beginning STPA analysis. After loading stpa/overview. When you need to establish what could go wrong and what must be prevented.
Skill Claude CodeCodex
STPA Step 2 - Model the control structure using hierarchical control-feedback diagrams in Graphviz/DOT format; After completing STPA Step 1. When you need to understand how control flows through a system. When identifying controllers, control actions, and feedback paths.
Skill Claude CodeCodex
STPA Step 3 - Identify Unsafe Control Actions (UCAs) using the 4-type analysis framework. After completing STPA Step 2. When analyzing control actions for potential safety issues. When you need to systematically identify what could go wrong with each control action.
Skill Claude CodeCodex
STPA Step 4 - Identify Loss Scenarios by tracing causal pathways back through control loops to understand why UCAs might occur. After completing STPA Step 3. When you need to understand WHY unsafe control actions might happen. When developing recommendations and mitigations.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: