A procedure for launching the olas-website Next.js development server and checking the result in a browser. Next.js is a framework for building web applications with JavaScript and React.

Use it when running the olas-website locally, checking a change in a browser, or taking a screenshot of a page.

It makes local changes easier to verify by starting the correct development environment, finding the actual local address, and checking the rendered page.

How valory-xyz/olas-website configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to valory-xyz/olas-website. 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/valory-xyz/olas-website/main/.claude/skills/run-dev/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/valory-xyz/olas-website

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/valory-xyz/olas-website/run-dev/github.svg)](https://agentmods.dev/skills/valory-xyz/olas-website/run-dev)

Your own site

<a href="https://agentmods.dev/skills/valory-xyz/olas-website/run-dev"><img src="https://agentmods.dev/badge/skills/valory-xyz/olas-website/run-dev/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/valory-xyz/olas-website/run-dev"><img src="https://agentmods.dev/badge/skills/valory-xyz/olas-website/run-dev.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 43 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 953 The whole file, excluding the scripts and references it only reads on demand.

Security scan C 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.

Model	Per session	Once invoked
Fable 5.1	$0.00043	$0.00953
Opus 5.5	$0.00017	$0.00381
Sonnet 5.5	$0.00009	$0.00191
Haiku 4.5	$0.00004	$0.00095

Measured 2mo ago against content hash 935acb62a6d2, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade C, and why

run-dev scanned grade C 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 2mo 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.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

caches aggressively in dev. Fix: `rm -rf .next/cache/images`, then hard

Makes network callslowCapability

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

for i in $(seq 1 60); do curl -sf http://localhost:<port> >/dev/null && break; sleep 1; done

.claude/skills/run-dev/SKILL.md · 89 lines

How it starts

The opening of the file, as written. The whole thing — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Run the olas-website dev server

Node/yarn setup (required first)

node/yarn are NOT on PATH in non-interactive shells — Node is managed by nvm, which only initializes in interactive shells. Load nvm and let it resolve the version from the repo's .nvmrc (single source of truth — nvm install installs it if missing, otherwise just switches to it):

source "$HOME/.nvm/nvm.sh" && nvm install

yarn (classic, /usr/local/bin/yarn) works once node is on PATH.

Note: shell state does not persist between Bash tool calls — re-source nvm in each call that needs node/yarn, or chain commands with && in one call.

Launch

yarn dev   # run in background

Next may not get port 3000 (it auto-falls back to 3001, 3002, …). Read the actual port from the dev-server output line - Local: http://localhost:<port> before navigating. If another process holds a port, it isn't ours — leave it.
Ready in ~3s. Poll, don't sleep — and note macOS has no timeout command:

for i in $(seq 1 60); do curl -sf http://localhost:<port> >/dev/null && break; sleep 1; done

.env.local exists in the repo root and is sufficient for local rendering. Homepage metrics come from Vercel Blob snapshots at build/ISR time, so pages render fine even if some subgraph env vars are missing.

Stop

lsof -ti:<port> -sTCP:LISTEN | xargs -r kill

Screenshots / driving the app

Pick a driver in this order:

chromium-cli if installed (which chromium-cli) — headless, no extension needed.
claude-in-chrome skill (Chrome extension MCP tools): tabs_context_mcp → navigate → javascript_tool to scroll → computer screenshot with save_to_disk: true. There is no upfront check for the extension — invoking the skill is the check: it fails (or walks through setup) when the extension isn't installed/connected. If that happens, tell the user they can install it from https://claude.ai/chrome, and fall back to option 3 meanwhile.
Playwright fallback (no extension, no chromium-cli): npx playwright with a one-off script — chromium.launch() → page.goto(devUrl) → page.screenshot(). First run downloads a browser (npx playwright install chromium), so prefer 1/2 when available.

Read the full file on GitHub · 89 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.

2mo ago First seen · 89 lines · 43 tokens per session scan C 935acb62a6d2

Subscribe to this mod's changes

run-dev is a skill published in the GitHub repository valory-xyz/olas-website (2 stars, last pushed 2d ago), licensed Apache-2.0. It adds 43 tokens to every session and 953 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Browser automation Frontend agents ai ai-agents crypto olas

run-dev next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
run-devvalory-xyz/olas-website	Skill	2	2d ago	Apache-2.0	C	43	Claude Code
browsebrowserbase/stagehand	Skill	26k	yesterday	MIT	A	109	Claude Code
playwrightopenai/openai-agents-python	Skill	31k	yesterday	MIT	A	19	Claude Code, Codex
playwright-cliVoltAgent/voltagent	Skill	11k	2d ago	MIT	A	52	Claude Code
azure-messaging-webpubsub-javamicrosoft/skills	Skill	3.2k	yesterday	MIT	A	43	Claude Code
computer-usekirodotdev/KiroCrew	Skill	4.8k	yesterday	Apache-2.0	A	62	Claude Code, Codex
feature-demo-recordingkirodotdev/KiroCrew	Skill	4.8k	yesterday	Apache-2.0	A	90	Claude Code, Codex

olas-website: Skill for Claude Code