model-strategy

Solid

Multi-Modell Orchestrierung und Model-Switching Strategie. Score-basierte Modellauswahl, Cross-Agent-Delegation (Gemini, Codex, Ollama), Advisor-Pairing, Eskalations-Trigger, Berechtigungsmatrix und Kosten-Effizienz-Optimierung.

AI & Automation 2 stars 1 forks Updated yesterday MIT

Install

Quality Score: 81/100

Stars 20%

16

Recency 20%

100

Frontmatter 20%

70

Documentation 15%

100

Issue Health 10%

80

License 10%

100

Description 5%

100

Skill Content

<img src="banner.png" width="100%" alt="model-strategy banner"> # Model-Switching Strategie > Multi-Modell Orchestrierung: Score-basierte Modellauswahl, Cross-Agent-Delegation, Advisor-Pairing, Eskalations-Trigger und Kosten-Effizienz-Optimierung --- ## 1. Modell-Katalog ### Claude (Subagent-fähig via Agent-Tool) ``` Level 4 (Reviewer): Opus 4.8 — Advisor, Math-Review [nur User: /model, /advisor] Level 3 (Stratege): Opus 4.6 — Architektur, Konzepte [Subagent: model:"opus"] Level 3 (Kreativ): Fable 5 — Kreative Texte, Stories [Subagent: model:"fable"] Level 2 (Arbeitstier):Sonnet 4.6— Implementation, Debug [Subagent: model:"sonnet"] Level 1 (Schnell): Haiku 4.5 — Boilerplate, Formatting [Subagent: model:"haiku"] ``` ### Externe Agenten (Companion-Scripts / SSH) ``` Level 2-3: Gemini 3.5 pro — Recherche, wiss. Datenbanken [agy-companion CLI] Level 2: Gemini 3.5 flash— Schnelle Recherche [agy-companion CLI] Level 2-3: Codex 5.5 (GPT) — Code-Review, Code-Gen [codex-companion CLI] Level 2: Codex 4.5 (GPT) — Einfachere Code-Aufgaben [codex-companion CLI] ``` ### Lokale Modelle (Token-frei, 24/7) ``` Level 1-2: Ollama (Qwen 3.5:35b-a3b) — Haiku-bis-Sonnet-Niveau [<ollama-host>:11434] Aufruf: SSH + curl http://<ollama-host>:11434/v1/chat/completions Oder: Buddha-Delegation via BACH Control API :8081 ``` ### Erreichbarkeits-Matrix | Modell | LLM-startbar | Aufrufweg | Einschränkungen | |--------|-...

Details

Author: ellmos-ai
Repository: ellmos-ai/skills
Created: 4 months ago
Last Updated: yesterday
Language: Python
License: MIT

Integrates with

Anthropic · AI Ollama · AI SQLite · Database

Similar Skills

Semantically similar based on skill content — not just same category

Testing & QA Listed

co-pilot-loop

Cooperative Multi-Agent Coordination Loop. Spawns a lightweight companion sub-agent (Gemini 3.5 Flash Low) to perform spec discovery, planning, and implementation. The primary agent (Claude) acts as the QA Director, answering Gemini's questions, approving the spec/plan, and running verification tests. Assumes the superpowers plugin is installed in the target repository.

4 Updated yesterday

AI & Automation Listed

stuntdouble

Your frontier model is the star — stop making it do the stunts. Two modes. ADVISOR: analyzes a task and recommends the best AI model (Anthropic/OpenAI/Google) with the exact switch command. ORCHESTRATOR: stays ON for the whole session (like a mode toggle) — every task gets scored, decomposed, and dispatched to "stunt double" workers running the cheapest capable model OF THE HOST'S OWN PROVIDER (Claude models in Claude Code, OpenAI models in Codex, Gemini models in Gemini CLI), with the star (strongest model) doing only the final close-up: verification. TRIGGER on /stuntdouble, "stuntdouble on/off", "which model should I use", "select the best model", "optimize model usage", "is this task worth Opus". DO NOT TRIGGER when the user asks about model pricing/APIs in their own application code.

1 Updated 1 weeks ago

AI & Automation Listed

advisor-executor

Cut the cost of Claude agents by splitting the work across models — a cheap executor with a strong advisor tool (advisor_20260301), or a strong orchestrator delegating to cheap workers (Managed Agents multiagent). Use when building or reviewing an Anthropic API agent and the ask involves cost, price, spend, budget, token bill, "too expensive", model choice/routing/escalation, using Fable/Opus to steer Sonnet/Haiku, planner-worker or coordinator-subagent designs, or getting frontier quality without frontier cost. Also covers the subscription equivalents (no API key) — stretching Claude Pro/Max usage limits in Claude Code via opusplan, advisor/scout subagents with per-agent model overrides, and the Agent SDK on a setup-token.

0 Updated 1 weeks ago