/makers-deploy
This skill deploys frontend and full-stack projects to EdgeOne Makers (Tencent EdgeOne). Trigger this skill whenever deployment is part of the task — whether as the primary intent or a secondary step. Examples: "deploy my app", "publish this site", "push this live", "create a
$ npx -y skills add tencentedgeone/edgeone-pages-skills --skill makers-deploy --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
- Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
- You can call itInvoke it directly when you want it.
- Slash command
/makers-deploy
Context preview
The summary Claude sees to decide when to auto-load this skill.
This skill deploys frontend and full-stack projects to EdgeOne Makers (Tencent EdgeOne). Trigger this skill whenever deployment is part of the task — whether as the primary intent or a secondary step. Examples: "deploy my app", "publish this site", "push this live", "create a
SKILL.md
makers-deploy.SKILL.mdname: edgeone-makers-deploy
description: >-
This skill deploys frontend and full-stack projects to EdgeOne Makers (Tencent EdgeOne).
Trigger this skill whenever deployment is part of the task — whether as the primary intent
or a secondary step. Examples: "deploy my app", "publish this site", "push this live",
"create a preview deployment", "deploy to EdgeOne", "ship to production",
"go live", "release", "publish a new version", "redeploy",
"上线", "发布", "发一版", "重新部署",
"搭建并部署", "开发并上线", "build and deploy", "create and deploy".
⚠️ Also trigger when any agent is about to execute `edgeone makers deploy` or `edgeone makers deploy`
commands — the skill contains critical rules for parsing deploy output and presenting access URLs.
Do NOT trigger for post-deployment runtime errors (e.g. CORS issues, 500 errors after deploy —
use edgeone-makers-dev for troubleshooting).
metadata:
author: edgeone
version: "2.2.0"
EdgeOne Makers Deployment Skill
Deploy any project to **EdgeOne Makers**.
⛔ Critical Rules (never skip)
1. **CLI version ≥ `1.6.0`** — reinstall if lower. Versions below `1.6.0` lack the non-interactive fixes (whoami fail-fast, `--json` output) and will hang in Agent/CI environments. Never proceed with an outdated version. 2. **Never truncate the deploy URL — this applies to EVERY mention** — `EDGEONE_DEPLOY_URL` includes query parameters (`?eo_token=...&eo_time=...`) required for access. Without them the page returns 401. Always output the **complete** URL with full query string. This rule applies to: the primary display, summary tables, footnotes, comparisons, code blocks, `present_files` calls — **every single occurrence** of the URL in your reply. Truncation is any removal of the `?` and everything after it.
❌ WRONG (truncated — will 401):
https://my-app-w9t0lxe8.edgeone.cool
✅ CORRECT (full URL):
https://my-app-w9t0lxe8.edgeone.cool?eo_token=abc123&eo_time=1234567890
**Self-check after writing your reply**: scan for every instance of the `.edgeone.cool` domain. Does each one include `?eo_token=`? If any doesn't, fix it NOW — the user will get a 401. 3a. **Prefer `--json` when running non-interactively** — in Agent/CI/headless contexts, always pass `--json` to `deploy` so the result is a single machine-readable line; no need to scrape colored/`\r`-animated stdout. See **Parse Deploy Output**. 3b. **Use `edgeone whoami` to check login status** — on CLI ≥ 1.6.0, `whoami` fails fast (exit 1) when not logged in instead of hanging. If it exits 0, the user is already logged in and `-t` is not needed. **Do NOT** check `cat .edgeone/.token` — CLI stores credentials in `~/.edgeone/<hash>` files, not a fixed `.token` path. 4. **⚠️ The deploy URL MUST be placed prominently at the very top of your reply** — once deployment finishes, the complete access URL is the core deliverable the user cares about most. You MUST: ① place it on the first line or in the first standalone block of your reply body; ② use a prominent format (e.g. a large heading + code block); ③ never bury the URL in the middle of a long paragraph where the user has to hunt for it. Example format:
🌐 Live URL: https://my-project-abc123.edgeone.cool?<auth_query_params>
Then append any other notes (console URL, caveats, etc.). 5. **Ask the user to choose China or Global site** before browser login. Never assume. (Token login via `edgeone login --token` auto-detects site, no need to ask.) 6. **Auto-detect the login method** — browser login in desktop environments, token login in headless/remote/CI environments. Follow the decision table below. 7. **After token login, ask if the user wants to save the token locally** for future use. 8. **Before triggering any browser popup (login / registration), explain the reason and the benefits to the user first** — never silently launch a browser window.
---
Environment Setup
Before executing **any** `edgeone` CLI command (install, login, deploy, etc.), set the following environment variable in the current shell session:
export PAGES_SOURCE=skills
Or prefix each command inline:
PAGES_SOURCE=skills edgeone makers deploy
This tells the platform that the deployment is triggered from an AI skill context.
---
Deployment Flow
Run these checks first, then follow the decision table:
# Check 0: Set environment variable (required before any edgeone command)
export PAGES_SOURCE=skills
# Check 1: CLI installed and correct version? (must be >= 1.6.0)
edgeone -v
# Check 2: Already logged in? (CLI >= 1.6.0 whoami fails fast, won't hang)
edgeone whoami
# If exit 0 → logged in, no -t needed
# If exit 1 → not logged in, need token or browser login
# Check 3: Project already linked?
cat edgeone.json 2>/dev/null
Decision Table
| CLI version | Login status | Action | |-------------|-------------|--------| | Not installed or < 1.6.0 | — | → Go to **Install CLI** | | `≥ 1.6.0` ✓ | Logged in (or token present) | → Go to **Deploy** | | `≥ 1.6.0` ✓ | Not logged in, has saved token | → Go to **Deploy with Token** (use saved token) | | `≥ 1.6.0` ✓ | Not logged in, no saved token, **interactive desktop** | → Go to **Login** (browser) | | `≥ 1.6.0` ✓ | Not logged in, no saved token, **non-interactive (Agent/CI/headless)** | → Ask user for a **token**; browser login is unavailable and `deploy` will fail fast with a token hint |
---
Install CLI
npm install -g edgeone@latest
Verify: `edgeone -v` — confirm output is `1.6.0` or higher. Retry installation if not. (Versions < 1.6.0 hang on `whoami`/login in non-interactive environments and lack `--json`.)
---
Login
0. Explain the registration/login step
Before triggering any login flow, explain to the user **why** this step is needed and **what** to expect. Do not silently launch a browser window.
Tell the user:
> You need to log in or register an EdgeOne Makers account. Here's what to ex
Read more
name: edgeone-makers-deploy description: >- This skill deploys frontend and full-stack projects to EdgeOne Makers (Tencent EdgeOne). Trigger this skill whenever deployment is part of the task — whether as the primary intent or a secondary step. Examples: "deploy my app", "publish this site", "push this live", "create a preview deployment", "deploy to EdgeOne", "ship to production", "go live", "release", "publish a new version", "redeploy", "上线", "发布", "发一版", "重新部署", "搭建并部署", "开发并上线", "build and deploy", "create and deploy". ⚠️ Also trigger when any agent is about to execute `edgeone makers deploy` or `edgeone makers deploy` commands — the skill contains critical rules for parsing deploy output and presenting access URLs. Do NOT trigger for post-deployment runtime errors (e.g. CORS issues, 500 errors after deploy — use edgeone-makers-dev for troubleshooting). metadata: author: edgeone version: "2.2.0"
EdgeOne Makers Deployment Skill
Deploy any project to **EdgeOne Makers**.
⛔ Critical Rules (never skip)
1. **CLI version ≥ `1.6.0`** — reinstall if lower. Versions below `1.6.0` lack the non-interactive fixes (whoami fail-fast, `--json` output) and will hang in Agent/CI environments. Never proceed with an outdated version. 2. **Never truncate the deploy URL — this applies to EVERY mention** — `EDGEONE_DEPLOY_URL` includes query parameters (`?eo_token=...&eo_time=...`) required for access. Without them the page returns 401. Always output the **complete** URL with full query string. This rule applies to: the primary display, summary tables, footnotes, comparisons, code blocks, `present_files` calls — **every single occurrence** of the URL in your reply. Truncation is any removal of the `?` and everything after it.
❌ WRONG (truncated — will 401):
https://my-app-w9t0lxe8.edgeone.cool
✅ CORRECT (full URL):
https://my-app-w9t0lxe8.edgeone.cool?eo_token=abc123&eo_time=1234567890
**Self-check after writing your reply**: scan for every instance of the `.edgeone.cool` domain. Does each one include `?eo_token=`? If any doesn't, fix it NOW — the user will get a 401. 3a. **Prefer `--json` when running non-interactively** — in Agent/CI/headless contexts, always pass `--json` to `deploy` so the result is a single machine-readable line; no need to scrape colored/`\r`-animated stdout. See **Parse Deploy Output**. 3b. **Use `edgeone whoami` to check login status** — on CLI ≥ 1.6.0, `whoami` fails fast (exit 1) when not logged in instead of hanging. If it exits 0, the user is already logged in and `-t` is not needed. **Do NOT** check `cat .edgeone/.token` — CLI stores credentials in `~/.edgeone/<hash>` files, not a fixed `.token` path. 4. **⚠️ The deploy URL MUST be placed prominently at the very top of your reply** — once deployment finishes, the complete access URL is the core deliverable the user cares about most. You MUST: ① place it on the first line or in the first standalone block of your reply body; ② use a prominent format (e.g. a large heading + code block); ③ never bury the URL in the middle of a long paragraph where the user has to hunt for it. Example format:
🌐 Live URL: https://my-project-abc123.edgeone.cool?<auth_query_params>
Then append any other notes (console URL, caveats, etc.). 5. **Ask the user to choose China or Global site** before browser login. Never assume. (Token login via `edgeone login --token` auto-detects site, no need to ask.) 6. **Auto-detect the login method** — browser login in desktop environments, token login in headless/remote/CI environments. Follow the decision table below. 7. **After token login, ask if the user wants to save the token locally** for future use. 8. **Before triggering any browser popup (login / registration), explain the reason and the benefits to the user first** — never silently launch a browser window.
---
Environment Setup
Before executing **any** `edgeone` CLI command (install, login, deploy, etc.), set the following environment variable in the current shell session:
export PAGES_SOURCE=skills
Or prefix each command inline:
PAGES_SOURCE=skills edgeone makers deploy
This tells the platform that the deployment is triggered from an AI skill context.
---
Deployment Flow
Run these checks first, then follow the decision table:
# Check 0: Set environment variable (required before any edgeone command) export PAGES_SOURCE=skills # Check 1: CLI installed and correct version? (must be >= 1.6.0) edgeone -v # Check 2: Already logged in? (CLI >= 1.6.0 whoami fails fast, won't hang) edgeone whoami # If exit 0 → logged in, no -t needed # If exit 1 → not logged in, need token or browser login # Check 3: Project already linked? cat edgeone.json 2>/dev/null
Decision Table
| CLI version | Login status | Action | |-------------|-------------|--------| | Not installed or < 1.6.0 | — | → Go to **Install CLI** | | `≥ 1.6.0` ✓ | Logged in (or token present) | → Go to **Deploy** | | `≥ 1.6.0` ✓ | Not logged in, has saved token | → Go to **Deploy with Token** (use saved token) | | `≥ 1.6.0` ✓ | Not logged in, no saved token, **interactive desktop** | → Go to **Login** (browser) | | `≥ 1.6.0` ✓ | Not logged in, no saved token, **non-interactive (Agent/CI/headless)** | → Ask user for a **token**; browser login is unavailable and `deploy` will fail fast with a token hint |
---
Install CLI
npm install -g edgeone@latest
Verify: `edgeone -v` — confirm output is `1.6.0` or higher. Retry installation if not. (Versions < 1.6.0 hang on `whoami`/login in non-interactive environments and lack `--json`.)
---
Login
0. Explain the registration/login step
Before triggering any login flow, explain to the user **why** this step is needed and **what** to expect. Do not silently launch a browser window.
Tell the user:
> You need to log in or register an EdgeOne Makers account. Here's what to ex
Official AI Agent Skills for developing and deploying projects on EdgeOne Makers.
Repo: tencentedgeone/edgeone-pages-skills
Other skills on edgeone-makers-tools.
- /makers-agents
This skill guides building AI agent endpoints on EdgeOne Makers — five framework routes (DeepAgents, LangGraph, CrewAI, OpenAI Agents SDK, Claude Agent SDK), platform-injected `context.store` / `context.tools` / `context.sandbox`, conversation_id dual-channel routing, SSE
Open skill - /makers-cli
EdgeOne Makers CLI command reference. Use when running edgeone CLI commands for dev, build, deploy, env management.
Open skill - /makers-cloud-functions
EdgeOne Makers Cloud Functions — Node.js, Go, and Python runtimes. Use when building server-side APIs, Express/Koa patterns, or backend logic.
Open skill - /makers-edge-functions
V8-based lightweight edge functions on EdgeOne Makers. Covers routing, KV storage access, request/response handling, and environment variables at the edge.
Open skill - /makers-env-adaption
Environment-specific adaptation rules for EdgeOne Makers Skills running in sandboxed or restricted AI coding environments (e.g. WorkBuddy). Trigger when: the user is working in WorkBuddy, a sandboxed IDE, or any non-interactive/CI environment where CLI commands may hang or
Open skill - /makers-middleware
Edge middleware for EdgeOne Makers — request interception, redirects, rewrites, auth guards, A/B testing, and header injection at the edge (V8 runtime).
Open skill

