A browser-based smoke test procedure for Hybro's user interface, meaning a quick check of the main experience after a change. It turns findings into Playwright tests, which are automated browser checks.

Use it after frontend or cross-service changes to check areas such as chat, agents, settings, sessions, and responsive behavior.

It catches broken screens, flows, responsive layouts, and console errors before they reach users, without changing the code during the inspection.

How hybroai/hybro configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to hybroai/hybro. 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/hybroai/hybro/main/.claude/skills/ui-smoke/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/hybroai/hybro

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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/hybroai/hybro/ui-smoke/github.svg)](https://agentmods.dev/skills/hybroai/hybro/ui-smoke)

Your own site

<a href="https://agentmods.dev/skills/hybroai/hybro/ui-smoke"><img src="https://agentmods.dev/badge/skills/hybroai/hybro/ui-smoke/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/hybroai/hybro/ui-smoke"><img src="https://agentmods.dev/badge/skills/hybroai/hybro/ui-smoke.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 72 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 938 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 1 finding. 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.

Model	Per session	Once invoked
Fable 5.1	$0.00072	$0.00938
Opus 5.5	$0.00029	$0.00375
Sonnet 5.5	$0.00014	$0.00188
Haiku 4.5	$0.00007	$0.00094

Measured 6d ago against content hash ed584778c45d, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

ui-smoke scanned grade A with 1 finding 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 6d 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.

Makes network callslowCapability

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

1. `curl -s <base>/api/v1/health` (or the root) answers. If not, say so and stop; do not start services unless asked.

.claude/skills/ui-smoke/SKILL.md · 45 lines

How it starts

The opening of the file, as written. The whole thing — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Hybro UI smoke walk

Arguments: $ARGUMENTS. First token that looks like a URL is the base URL; the rest is the focus area. Default base URL is the dev instance http://localhost:3100 (./scripts/hybro start --build, container hybro-dev). Never target the release instance on port 3000 unless the URL is given explicitly, and never use real provider keys or paid models unless told to.

Use the built-in browser (mcp__Claude_Browser__*). Prefer read_page and get_page_text for checks; screenshot only when layout or styling is the question. Record every observation as you go; do not fix anything until the walk is finished.

0. Preconditions

curl -s <base>/api/v1/health (or the root) answers. If not, say so and stop; do not start services unless asked.
Clear console messages, open <base>/.

1. Flows (skip ones outside the focus area, say which)

Overview /: loads without errors; provider banner present only when no provider is configured; banner is dismissible here.
Agents /agents: sample agents listed (Echo always; LLM-backed ones show "needs model" when unconfigured); health state visible; agent detail opens.
Chat /chat: quick-start templates render; submitting creates a room and navigates without a sign-in redirect. If no provider is configured: permanent banner and disabled composer.
Room: send a message, a turn appears; multi-agent reply grouped in one turn; activity strip collapses and expands; agent details open with history column still visible; no internal dispatch prompt ever shown in the stream; cancel in-flight works.
HITL: when an agent asks for input, the reply control appears, the answer is accepted, and the resolved interaction stays visible after reload.
Files: attach a small file and send; artifact shows a download control.
Networks /networks: graph renders registered and discovered agents; search, scope and selection survive reload.
Sessions /sessions: list renders; owner can approve a client session.
Manage /manage and MCP connect page: readable and copyable at 1280px and 375px; legacy management routes redirect.
Model provider settings: banner opens settings; saving a key clears the banner (use a dummy key only if the UI validates format, not a live call).
Errors: unknown route gives a 404 page; /rooms/<id>/anything returns 404, not a settings route.
Responsive: repeat Overview, Chat and a Room at 375x812 and 1280x800; no horizontal scroll, controls reachable.
Console: zero errors across the walk; list any warnings that repeat.

Read the full file on GitHub · 45 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.

6d ago First seen · 45 lines · 72 tokens per session scan A ed584778c45d

Subscribe to this mod's changes

ui-smoke is a skill published in the GitHub repository hybroai/hybro (18 stars, last pushed 3d ago), licensed Apache-2.0. It adds 72 tokens to every session and 938 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-11-01.

Browser automation E2E testing Multi-Agent Testing a2a-protocol agent-orchestration ai-agents fastapi hybrid-agents

ui-smoke next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
ui-smokehybroai/hybro	Skill	18	3d ago	Apache-2.0	A	72	Claude Code
webapp-testingHmbown/CodeWhale	Skill	41k	6d ago	MIT	A	32	Claude Code, Codex
qaSethGammon/Citadel	Skill	927	yesterday	MIT	A	56	Claude Code
browser_scriptstubecreate/tubecli	Skill	174	yesterday	MIT	A	0	Claude Code, Codex
chrome-automationaAAaqwq/AGI-Super-Team	Skill	114	yesterday	MIT	A	34	Claude Code
run-dashboard-e2e-local-changesBlackBeltTechnology/pi-agent-dashboard	Skill	341	2d ago	MIT	A	42	Claude Code, Codex
browser-checkjjanczur/tyran	Skill	79	yesterday	Apache-2.0	A	58	Claude Code

hybro: Skill for Claude Code