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 SunflowersLwtech/sunkit --skill web-deploygit clone --depth 1 https://github.com/SunflowersLwtech/sunkitWrote 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/sunflowerslwtech/sunkit/web-deploy)<a href="https://agentmods.dev/skills/sunflowerslwtech/sunkit/web-deploy"><img src="https://agentmods.dev/badge/skills/sunflowerslwtech/sunkit/web-deploy/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/sunflowerslwtech/sunkit/web-deploy"><img src="https://agentmods.dev/badge/skills/sunflowerslwtech/sunkit/web-deploy.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00177 | $0.01756 |
| Opus 5.5 | $0.00071 | $0.00702 |
| Sonnet 5.5 | $0.00035 | $0.00351 |
| Haiku 4.5 | $0.00018 | $0.00176 |
Grade A, and why
web-deploy 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 yesterday.
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.
How it starts
The opening of the file, as written. The whole thing — 107 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Deploying web apps (Vercel · Render · GitHub Actions)
Pushing to main is often the production deploy
In a bootstrapped repo, .github/workflows/deploy.yml fires on: push: branches: [main]
and ends in vercel deploy --prebuilt --prod. There is no separate "release" step to
forget — the push is the release. Look before you push:
ls .github/workflows/ && sed -n '1,10p' .github/workflows/deploy.yml
Ask a human first for: rotating or replacing a deploy token, deleting or overwriting an existing env var, the first production deploy of a project nobody has shipped yet, and anything that changes what a live domain serves. These are outward-visible and cheap to ask about, expensive to undo.
Find the repo's one deploy path — do not guess it
Three shapes, mutually exclusive on purpose:
| Shape | How you tell | How it deploys |
|---|---|---|
| GitHub Actions → Vercel | .github/workflows/deploy.yml and vercel.json with git.deploymentEnabled=false |
push to main, or gh workflow run deploy.yml. Watch the Actions run, not the Vercel dashboard |
| Vercel CLI, prebuilt | .vercel/project.json present, no deploy workflow |
vercel build --prod then vercel deploy --prebuilt --prod --scope <slug>. It uploads the local output, so commit first or you ship uncommitted work. It also replaces the whole live site with your tree: when other sessions deploy the same project, build from a fresh origin/main and first compare the live copy of every file git diff --name-only <live commit> origin/main -- <site dir> names with your tree; a live file you don't have means someone shipped after your base, and deploying rolls it back |
| Render, GHCR image | the service is in services.json |
push to main; the repo's own CI builds and pushes the image. bash scripts/render.sh release <service> is the fallback for when CI is unavailable |
git.deploymentEnabled=false is not decoration: it stops Vercel's native Git integration
from racing the workflow. Where it is missing, one push produces two deployments and
"which build is live" stops having an answer.
What ships with it
7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- yesterday First seen · 107 lines · 177 tokens per session scan A 6d517ec64928
web-deploy is a skill published in the GitHub repository SunflowersLwtech/sunkit (13 stars, last pushed today), licensed Apache-2.0. It adds 177 tokens to every session and 1,756 once invoked, about $0.0007 per session on Opus 5.5. 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 2026-11-12.
web-deploy compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| web-deploySunflowersLwtech/sunkit | Skill | 13 | today | Apache-2.0 | A | 177 | Claude Code |
| seed-snapshot-releasedaangn/seed-design | Skill | 1.2k | today | Apache-2.0 | A | 19 | Claude Code, Codex |
| github-actions-expertroedyrustam/vibes-plug | Skill | 126 | 2d ago | MIT | A | 35 | Cursor |
| gearsimota/agent-skills | Skill | 99 | 2d ago | MIT | A | 32 | Claude Code, Codex |
| github-release-managementfrankxai/claude-skills-library | Skill | 47 | today | MIT | A | 43 | Claude Code |
| github-masterMiaoge-Ge/coding-agent-skills | Skill | 7 | 5mo ago | MIT | A | 85 | Claude Code |
| github-workflow-automationdxkjuanjuan/ui-forge | Skill | 3 | 3mo ago | none | A | 27 | Claude Code, Codex |