vLLM Semantic Router is a programmable routing layer that chooses or combines language models for each request in a system using multiple models and types of computing infrastructure. It helps teams route inference by signals such as user preferences, application policies, quality, cost, latency, privacy, and safety requirements. The catalogue skills and instructions support configuring and operating this model-routing system.
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 vllm-project/semantic-router --skill routing-calibrationgit clone --depth 1 https://github.com/vllm-project/semantic-routerWrote 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/vllm-project/semantic-router/routing-calibration)<a href="https://agentmods.dev/skills/vllm-project/semantic-router/routing-calibration"><img src="https://agentmods.dev/badge/skills/vllm-project/semantic-router/routing-calibration/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/vllm-project/semantic-router/routing-calibration"><img src="https://agentmods.dev/badge/skills/vllm-project/semantic-router/routing-calibration.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 | $0.00027 | $0.00164 |
| Opus 5 | $0.00014 | $0.00082 |
| Sonnet 5 | $0.00005 | $0.00033 |
| Haiku 4.5 | $0.00003 | $0.00016 |
Grade A, and why
routing-calibration 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 2d 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.
What it actually says
Routing calibration
Use tools/dev/router-calibration/router_calibration_loop.py and the versioned
probe schema in config/schemas/recipe-probes-v1.schema.json. Keep recipe
changes in the local checkout; remote systems are validation mirrors only.
Treat endpoint credentials, hostnames, and private fleet details as secrets. Before a live run, validate the probe manifest and identify the exact recipe, backend, and rollback point. Record the evaluated inputs, outputs, and observed quality rather than inferring success from process health.
Run make check, then make verify DOMAIN=maintained-recipes for live
recipe verification.
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.
- 2d ago Changed · -77 lines · -26 tokens per session e716b9ddef5a
- 11d ago First seen · 96 lines · 53 tokens per session scan A 5c26e22a78b2
routing-calibration is a skill published in the GitHub repository vllm-project/semantic-router (5,717 stars, last pushed today), licensed Apache-2.0. It adds 27 tokens to every session and 164 once invoked, about $0.0001 per session on Opus 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 2026-08-30.
Other skills, from other repositories
mcp-harness
Use this harness to validate Agentweaver's MCP protocol surface, capture the complete tools/call request/response evidence, and emit a normalized agentweaver.persona-judge-verdict/v1 JSON verdict. It is for MCP end-to-end validation, MCP tool-contract regression checks, and investigation of an MCP-reported issue; use…
test-strategy-reproduction
Turn reported behavior into reliable repros, risk-based tests, and clear acceptance evidence.
systematic-debugging
4-phase root cause debugging: understand bugs before fixing.
cli-eval
Create and run evaluation suites, watch live benchmark progress, view scorecards, compare model performance, and integrate eval runs with CI workflows from the CLI.
experimental-code-coverage-local-debugger
Runs code coverage locally via Universal Test Runner (UTR) or helper scripts, mimicking LUCI trybots. Activate when CQ tryjobs fail or underreport coverage, to test local GN/recipe repairs before uploading, or to debug hermetic crashes.
adversarial-reviewer
Adversarial code review that assumes bugs exist and hunts for them. Use when asked to review code, find bugs, audit for correctness, stress-test a PR, or when someone says "tear this apart" or "what's wrong with this". Give no benefit of the doubt — every line is guilty until proven innocent.