Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add oyi77/1ai-skills --skill model-routergit clone --depth 1 https://github.com/oyi77/1ai-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/oyi77/1ai-skills/model-router)<a href="https://agentmods.dev/skills/oyi77/1ai-skills/model-router"><img src="https://agentmods.dev/badge/skills/oyi77/1ai-skills/model-router/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/oyi77/1ai-skills/model-router"><img src="https://agentmods.dev/badge/skills/oyi77/1ai-skills/model-router.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00039 | $0.03882 |
| Opus 5.5 | $0.00016 | $0.01553 |
| Sonnet 5.5 | $0.00008 | $0.00776 |
| Haiku 4.5 | $0.00004 | $0.00388 |
Grade A, and why
model-router scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 301 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Model Router
When to Use
Trigger phrases:
-
"model router"
-
"When managing multiple LLM providers (OpenAI, Anthropic, Google, etc"
-
"When optimizing for cost vs quality tradeoffs"
-
"When implementing fallback chains for reliability"
-
When managing multiple LLM providers (OpenAI, Anthropic, Google, etc.)
-
When optimizing for cost vs quality tradeoffs
-
When implementing fallback chains for reliability
-
When routing by task type (code to Claude, vision to Gemini, etc.)
When NOT to Use
- For single-provider setups (just use that provider SDK)
- For local-only inference (use Ollama skills)
Overview
Model routing is the intelligence layer between your application and the diverse landscape of large language model providers. Rather than hardcoding a single provider, a model router evaluates each request against dynamic criteria — task type, required capability, latency budget, cost ceiling, and current provider health — to select the optimal model at runtime. This transforms LLM integration from brittle point-to-point connections into a resilient, cost-aware system.
The core problem routing solves is that no single model dominates every dimension. Claude excels at code generation and structured reasoning, GPT-4o handles creative writing and multimodal tasks, Gemini-2.5-flash delivers the lowest latency for high-throughput workloads, and open-source models offer cost advantages for batch processing. Without routing, teams either overspend on premium models for trivial tasks or degrade quality by using cheap models everywhere. A router captures these trade-offs as programmable rules.
Production-grade routers implement three layers: static routing (model-per-task mapping), dynamic routing (cost/latency-aware selection with real-time metrics), and fallback chains (primary → secondary → tertiary on failure). Each layer adds resilience. When OpenAI experiences an outage, traffic shifts to Anthropic. When latency spikes on Claude, the router demotes it in favor of Gemini without any developer intervention.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 301 lines · 39 tokens per session scan A 20aac3de7497
model-router is a skill published in the GitHub repository oyi77/1ai-skills (17 stars, last pushed yesterday), licensed MIT. It adds 39 tokens to every session and 3,882 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
execution-hygiene
Runs computations so they finish, reproduce and fit the machine that will re-execute them, detaching long jobs through computejob instead of blocking or sleeping shells, checkpointing so reruns resume, fixing seeds and capping BLAS and OpenMP threads, streaming large tables with explicit dtypes, writing outputs…
generate-image
Generates or edits images with the native generateimage tool (Nano Banana Pro through Ace or a Gemini key, GPT Image 2 through an OpenAI key), for illustrations, cover art, graphical abstracts, conceptual visuals, photo-style figures and edits of existing images, saved straight into the workspace and checked before…
lynkr
Universal LLM gateway with intelligent routing, Graphify code intelligence, Distill compression, routing telemetry, Code Mode, and 12+ provider support. 60-80% cost reduction for Claude Code, Cursor, and Codex.
model-strategy
Rollenbasierte Multi-Modell-Strategie. Modellkatalog liegt in einer lokalen Konfigurationsdatei statt im Skill-Text -- Rollen (Denker/Entscheider, Dauer-Worker, Entwurfsgenerator, Reviewer/Abnahme, Eskalation), Reasoning-Effort als eigene Achse, Abnahmepflicht für Entwurfsmodelle, Anbieterwechsel für unabhängige…
skypilot-multi-cloud-orchestration
Multi-cloud orchestration for ML workloads with automatic cost optimization. Use when you need to run training or batch jobs across multiple clouds, leverage spot instances with auto-recovery, or optimize GPU costs across providers.
memory-rag-poisoning-detection
Detect persistent-memory poisoning and RAG/knowledge-base contamination via dialogue testing. Use when the agent has long-term memory, a user/preference store, or writes to a retrievable knowledge base.