static-deploy
Static-deploy uploads HTML, CSS, JS, and other static assets from a workspace directory to a Cloudflare Pages site at `<project-slug>.nexu.space`. Use it when a user requests deployment, publishing, or going live with a static website, single-page app, landing page, or documentation site. The script automatically injects cache headers to reduce stale asset issues on redeployments and handles both initial deploys and updates to existing sites.
git clone --depth 1 https://github.com/nexu-io/nexu /tmp/static-deploy && cp -r /tmp/static-deploy/skills/nexubot/static-deploy ~/.claude/skills/static-deploySKILL.md
# Static Deploy
Deploy static files to `<project-slug>.nexu.space` via Wrangler + Cloudflare Pages.
The deploy script stages a temporary copy of the site and injects a `_headers`
file (if needed) with cache revalidation rules for HTML/CSS/JS to reduce stale
asset issues after redeploys. Source files are not modified.
## What it deploys
Any static files: HTML, CSS, JS, images, fonts, etc. Common use cases:
- Single-page apps (React, Vue, etc. — deploy the `dist/` or `build/` folder)
- Landing pages, portfolios, documentation sites
- Quick prototypes or demos
- Any folder with an `index.html`
## Usage
1. Identify the directory containing files to deploy (must have an `index.html`)
2. Derive a project-slug from context (e.g., "family budget" → `family-budget`)
- Lowercase alphanumeric + hyphens only, max 63 characters
- Reuse the same slug for redeployments
- Ask user to confirm if ambiguous
3. Run deploy with the inbound structured context:
```bash
"$SKILL_DIR/scripts/deploy.sh" <project-slug> <directory> <agent-id> <chat-id> [thread-id] [message-ref] [account-id] [channel] [chat-type] [sender-ref]
```
`agent-id` is required. The script uses it to resolve the corresponding `botId`
from `nexu-context.json` when recording deployment artifacts.
`chat-id` is the raw inbound chat identifier from the message context.
- DM example: `U0AHLMC6C8G`
- Channel example: `C0AJKG60H6D`
`chat-type` tells the script whether to send that id to Nexu as `user:<chat-id>`
or `channel:<chat-id>`. The script does that translation; do not concatenate
those prefixes yourself.
`thread-id` is optional — pass it when deploying from within a thread context.
4. Parse the JSON output and report to user:
- Brief summary of what was deployed
- Live URL from the `url` field
- **Important:** Tell the user that Cloudflare Pages propagation takes ~3 minutes. The URL may not work immediately after the first deploy. For redeployments, it's usually faster.
## Example reply
> Deployed! Your site is live at https://family-budget.nexu.space
>
> Note: First-time deploys take ~3 minutes to propagate on Cloudflare. If you see a "not found" page, wait a few minutes and refresh.
## Rules
- Never read, echo, or log `SKILL_API_TOKEN`, `CLOUDFLARE_API_TOKEN`, or `CLOUDFLARE_ACCOUNT_ID`
- Never pass credentials as command arguments
- Always use the bundled `deploy.sh` — do not call Cloudflare API directly
- Do not set `DEPLOY_BACKEND` or any other env overrides — the script handles everything
- Always pass the caller `<agent-id>` as arg #3 to `deploy.sh`
- Always pass raw inbound `<chat-id>` as arg #4
- Never pass `<agent-id>` as `<chat-id>`; they are different values
- If deploying from a thread, pass `<thread-id>` as arg #5
- Always pass `<channel>` and `<chat-type>` when available so Nexu can resolve the correct session deterministically
- Do not try to look up or persist a canonical session key in this skill; Nexu resolves it server-side
- Do not hand-edit source files just to add cache-busting query params; the deploy script handles cache revalidation via a staged `_headers` file
- If the script fails, show the `message` field from the JSON error to the user
- Cloudflare credentials are fetched at runtime via the scoped secrets API using `SKILL_API_TOKEN` from env — do not attempt to read or inject them manually
## Output Format
Success:
```json
{
"status": "success",
"url": "https://<slug>.nexu.space",
"deployment_url": "https://<id>.<slug>.pages.dev",
"files_total": 10,
"files_uploaded": 2,
"files_cached": 8
}
```
Error:
```json
{
"status": "error",
"message": "Description of what went wrong"
}
```Use the ClawHub CLI to search, install, update, and publish agent skills from clawhub.com. Use when you need to fetch new skills on the fly, sync installed skills to latest or a specific version, or publish new/updated skill folders with the npm-installed clawhub CLI.
Delegate coding tasks to Codex, Claude Code, or Pi agents via background process. Use when: (1) building/creating new features or apps, (2) reviewing PRs (spawn in temp dir), (3) refactoring large codebases, (4) iterative coding that needs file exploration. NOT for: simple one-liner fixes (just edit), reading code (use read tool), thread-bound ACP harness requests in chat (for example spawn/run Codex or Claude Code in a Discord thread; use sessions_spawn with runtime:"acp"), or any work in ~/clawd workspace (never spawn agents here). Claude Code: use --print --permission-mode bypassPermissions (no PTY). Codex/Pi/OpenCode: pty:true required.
Use when the user says "check Datadog", "查 Datadog", "查日志", "check logs", "crash logs", "查 crash", "gateway crash", "查告警", "check alerts", "check metrics", or needs to investigate production issues via Datadog Logs API.
|
Send feedback to the Nexu team. Use when the user says /feedback followed by their message.
|
|
|