nexus-gateway: Skill for Claude Code

.claude/skills/smoke-gateway/SKILL.md

smoke-gateway is a skill for Claude Code from AlphaBitCore/nexus-gateway. It costs 234 tokens per session (3,467 once invoked), scanned B, original, Apache-2.0.

A broad automated test for an AI Gateway and its Control Plane, the service that manages gateway settings. It tests catalog models, request styles, caching, routing, database events, and monitoring counters, with an option to test the Responses API too.

In plain words
What is it for?
Use it after gateway or shared-code changes, when checking all provider models, or before declaring a gateway problem fixed. It can also test routing rules and Responses API behavior across the model catalog.
Why use it?
It checks many connected parts of the gateway after a change instead of relying on one example request. When configured for self-healing, it can investigate failures, change code, rebuild, restart, and retry.

Skill for Claude Code

Written for Claude Code: user-invocable in frontmatter. Also seen: mentions CLAUDE.md; positional $N argument.

This is AlphaBitCore/nexus-gateway's own configuration. It tells Claude Code how to work on nexus-gateway itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything nexus-gateway configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is go test -race -count=1 ./internal/... 2>&1.

Reuse

Borrowing it

Nothing to install: this file belongs to AlphaBitCore/nexus-gateway. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/AlphaBitCore/nexus-gateway/main/.claude/skills/smoke-gateway/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/AlphaBitCore/nexus-gateway

Made for: Claude Code.

Wrote 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.

agentmods badge for smoke-gateway

README.md
[![agentmods](https://agentmods.dev/badge/skills/alphabitcore/nexus-gateway/smoke-gateway/github.svg)](https://agentmods.dev/skills/alphabitcore/nexus-gateway/smoke-gateway)
Your own site
<a href="https://agentmods.dev/skills/alphabitcore/nexus-gateway/smoke-gateway"><img src="https://agentmods.dev/badge/skills/alphabitcore/nexus-gateway/smoke-gateway/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.

agentmods 80×15 button for smoke-gateway

Your own site · 80×15
<a href="https://agentmods.dev/skills/alphabitcore/nexus-gateway/smoke-gateway"><img src="https://agentmods.dev/badge/skills/alphabitcore/nexus-gateway/smoke-gateway.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 234 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,467 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 2 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00234 $0.03467
Opus 5 $0.00117 $0.01733
Sonnet 5 $0.00047 $0.00693
Haiku 4.5 $0.00023 $0.00347

Measured 10d ago against content hash 2f641b847041, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade B, and why

smoke-gateway scanned grade B with 2 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 10d 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.

Asks for rootmediumPrivilege escalation

A mod that escalates privileges can change anything on the machine, not only the project.

counts, (c) `sudo systemctl is-active nexus-ai-gateway` on prod, (d)

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -fsS http://localhost:3050/healthz
.claude/skills/smoke-gateway/SKILL.md · 289 lines

How it starts

The opening of the file, as written. The whole thing — 289 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Smoke Gateway

Full smoke test for the AI Gateway. Wraps tests/scripts/smoke-gateway.py and adds a self-healing loop: when tests fail, Claude investigates, fixes code, rebuilds, restarts the service, and reruns until green (or reports a root-cause it cannot auto-fix).

When to use

  • User types /smoke-gateway <vk> or asks to "smoke test all models", "verify gateway providers", "run full gateway test".
  • After any change to packages/ai-gateway/ or packages/shared/.
  • Before declaring a gateway bug fixed.

Inputs

Arg Required Notes
<vk> yes Virtual key (nvk_…).
--routing no Also run P4 per-rule routing tests.
--responses no Also run P3R: full per-model /v1/responses suite for every catalog model (non-stream + SSE + multi-round cache, mirrors P3 chat-completions depth). Models in the OpenAI native-support prefix list (gpt-5.x, gpt-4o, gpt-4.1, o1/o3/o4) exercise same-shape passthrough; every other catalog model exercises the E56 cross-format canonical bridge (Responses → canonical chat-completions → target wire → response → Responses-shape on egress). The cross-format guard + structured outputs + reasoning matrix still live in /test-openai-responses.
--messages no Also run P3A: full per-model /v1/messages (Anthropic ingress) suite. Same depth as P3. Native (Anthropic targets) and cross-format (canonical bridge to OpenAI / Gemini / others) both tested. Cached-token field: usage.cache_read_input_tokens.
--gemini no Also run P3G: full per-model /v1beta/models/{m}:generateContent (Gemini ingress) suite. Same depth as P3. Native (Gemini/Vertex targets) and cross-format both tested. Cached-token field: usageMetadata.cachedContentTokenCount.
--all-ingress no Umbrella: turn on --responses + --messages + --gemini in one go (every public ingress exercised at P3 depth). Forces cache test ON per binding feedback_cache_mandatory_all_ingress — every ingress must run the cache arm; if you also pass --no-cache it is ignored with a warning. Use this flag for full-surface validation after any ingress, codec, or canonical-bridge change.
--no-embeddings no Skip P3E embeddings phase (useful for quick chat-only smokes).
--all-upstream no Force all fixture-mode phases to real upstream (forward-compat hook for E63/E64/E66).
--no-stream no Skip SSE phase.
--no-cache no Skip 2-turn cache phase (ignored when --all-ingress is also set — see binding).
--cache-rounds N no Read rounds per cache test (default 3); shows positive ROI.
--fix-deprecated no P8: auto-delete catalog models that fail non-stream in P3.
--concurrent no Run P3.5 concurrent test (N=5 parallel).
--models m1,m2 no Restrict to specific models.
--timeout N no Per-request timeout in seconds (default 90).

Read the full file on GitHub · 289 lines

Changes

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.

  1. 10d ago First seen · 289 lines · 234 tokens per session scan B 2f641b847041

Subscribe to this mod's changes

smoke-gateway is a skill published in the GitHub repository AlphaBitCore/nexus-gateway (23 stars, last pushed yesterday), licensed Apache-2.0. It adds 234 tokens to every session and 3,467 once invoked, about $0.0012 per session on Opus 5. A static security scan graded it B with 2 findings (asks for root, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories