More screenshots
Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+22 moreClaude CodeCursorCodex CLIMuseDots
OpenClaw
Manus
Grok
Grok Bot+18 agentsDynamic Model Router — 3-Tier FinOps Routing Matrix
3-tier routing matrix (Flash-Lite vs Flash vs Pro/Sonnet) that slashes LLM API costs by 75-85% automatically.
$26
Dynamic Model Router — 3-Tier FinOps Routing Matrix
Example session with this skill installed
Analyze my 50 most recent agent task logs and show the cost comparison between default Claude 3.5 Sonnet vs 3-Tier Dynamic Model Routing.
- Read your context and instructions
- Compiled the dynamic model router
FinOps Model Routing Simulation (50 Tasks Analyzed)
| Tier Category | Task Distribution | Default Model (Sonnet 3.5) | Optimized Routing Matrix | Cost Reduction |
|---|---|---|---|---|
| Tier 1: Low (Regex, format, typos) | 22 tasks (44%) | $3.30 | Gemini Flash-Lite: $0.08 | -97.5% |
| Tier 2: Medium (Scripts, HTML, QA) | 20 tasks (40%) | $9.00 | Gemini Flash: $0.90 | -90.0% |
| Tier 3: High (Deep architecture, RFC) | 8 tasks (16%) | $7.20 | Claude Sonnet: $7.20 | 0.0% (Quality locked) |
| TOTAL | 50 tasks | $19.50 | $8.18 | -58.1% NET |
Configured Routing Rule:
Installed zero-overhead routing hook in rules/model_routing.md.
Next Steps:
Enable automated router hook: python3 scripts/model_router.py --daemon.
Connects securely to your tools. The creator never sees your data.
About this skill
Dynamic Model Router — 3-Tier FinOps Routing Matrix
Tired of seeing your LLM API bill spiral out of control because heavy reasoning models are used for trivial tasks?
The Dynamic Model Router dynamically evaluates incoming prompts and dispatches them across a 3-tier compute hierarchy:
- Tier 1 (Low / Flash-Lite): Text edits, regex checks, typos, quick shell scripts (-95% token cost).
- Tier 2 (Medium / Gemini Flash): File analysis, standard Python modules, HTML generation, QA checklists (-85% cost).
- Tier 3 (High / Claude Sonnet & Pro): Deep multi-file refactoring, GPU architecture, formal RFCs (100% capacity reserved).
⚡ Key Features & Engineering Guardrails:
- Production-grade architecture compatible with Claude Code, Cursor, Windsurf, and Antigravity.
- Zero external tracking: 100% local execution preserving API token economics.
- Complete protocol documentation and test harnesses included.
📦 What's Included in the ZIP:
- Complete production source code (
scripts/and CLI tools) - Architectural references and protocols (
references/) - Production prompt templates & schemas
How to use
Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scanned
Verified clean 2 days ago
- Passed all security checks, Safe to install