Generate architectural proposals, rank them via tournament evaluation, and record ADRs.
npx skills add https://github.com/markusstrasser/skills --skill architect-markusstrasser
Please help me install this Agent Skill. Skill: Architect Source: https://github.com/markusstrasser/skills/tree/main/archive/architect Command: npx skills add https://github.com/markusstrasser/skills --skill architect-markusstrasser
Architectural decision-making often becomes slow and uncertain when teams juggle competing proposals, lack a structured ranking method, and struggle to record decisions. This Skill provides a repeatable flow that collects proposals from multiple providers, ranks them using tournament-based evaluation, and records decisions as ADRs to improve traceability.
Run a full workflow with a problem description to generate proposals, rank them via tournament evaluation, and record the final ADR.
High-intent search queries and answers about installing and using this skill.
Evidence-based architectural decisions are made by generating proposals from multiple LLM providers, ranking them via tournament-based evaluation, and recording the outcome as a structured ADR for traceability.
Comparing microservice patterns or event-sourcing vs. CQRS involves gathering proposals from multiple LLM providers and using a tournament-based evaluator to identify and rank the strongest architectural direction.
Generating and ranking architectural proposals requires submitting a problem description to collect options from multiple LLM providers like Gemini, Codex, and Grok, then ranking them through tournament-based evaluation.
Yes, automating architectural decision-making requires Python 3.10+, an API key, and access to the llmx CLI to generate proposals and execute the tournament-based ranking workflow.
Yes, you can evaluate data-store strategies by generating multiple proposals, ranking them using tournament-based evaluation, and refining the final choice into a recorded Architectural Decision Record (ADR) for governance.
Tournament-based evaluation for architectural decision records works by pitting proposals from multiple LLM providers against each other to systematically rank competing designs and identify the most well-justified option.
alibabaConfigure Higress AI Gateway as a model provider for Clawdbot/OpenClaw.
CygnusfearDocument engineering decisions as ADRs with rationale and decision logs.
andrueandersoncsIntegrate and orchestrate AI/LLM services with Effect-TS using type-safe providers and tooling.
lazygophersDesign tournament structures, ranking systems, and tactical strategies for game development.
