dropage-deploy

Deploy an HTML page to the internet and return a public URL. Use when the user asks to deploy, host, share, or publish an HTML file or a zip / tar / tar.gz / tgz archive containing an index.html. Triggers on phrases like "deploy this page", "host this online", "share this HTML", "upload to web", "ge

Deploy to Dropage

Upload an HTML file or a zip / tar / tar.gz / tgz archive to dropage.online and return a public URL. Default expiry is 1 hour; the user may also choose 6 hours, 24 hours, 7 days, or 14 days, and optionally have the page disabled after 1 or 10 visits. Uploaded files are retained until 30 days after upload, independently of public link expiry.

Platform Detection

Before running the upload command, detect the user's platform and use the appropriate curl command:

• Windows: curl.exe
• macOS/Linux: curl

Optional upload fields

Both are optional; omit to use defaults.

• expiry — one of 1h (default), 6h, 24h, 7d, 14d
• max_visits — 0 (default, unlimited), 1, or 10. Page is disabled after the limit is reached.

Daily quotas and user reminders

Expiry	Site-wide successful uploads/day	Successful uploads/IP/day
7d	50	5
14d	10	1

The tiers have separate quotas. Each resets at 00:00 UTC+8 on the next calendar day. Only successful uploads count, including records later expired or cleaned; deletion does not refund quota. The 1h, 6h, and 24h tiers have no extra daily quota. All tiers still follow per-IP rate limits and the site-wide 1,000 uploads/hour limit.

Before a 7-day or 14-day upload, tell the user the selected tier's site-wide and per-IP daily limits and reset time. Use the requested expiry; if it is unspecified, use 1 hour. Never silently shorten a requested expiry to bypass an exhausted quota. Existing uploaded pages keep their expiration times; use the five current values for new uploads. The API still accepts legacy 10m, 3h, and 12h for old clients. A successful legacy response adds warning: {code: "LEGACY_EXPIRY", message: "...", skill_url: "https://dropage.online/dropage-deploy.md"}. Report that the upload succeeded, ask the user to update their installed skill from skill_url, and do not retry the successful upload.

Steps

1. Identify the file to deploy. Accept a single .html file or an archive (.zip, .tar, .tar.gz, .tgz) whose root contains index.html.

   Archive packaging requirements:

   • Supported formats: .zip, .tar, .tar.gz, .tgz (NOT .7z, .rar, .tar.bz2, .tar.xz)
   • index.html must be at the root level of the archive, not inside a subfolder
   • Correct: site.zip → index.html, style.css, js/
   • Wrong: site.zip → my-site/index.html, my-site/style.css
   • If all entries share a single top-level directory (e.g. my-site/), it is auto-stripped during extraction
   • Max 50 files per archive
   • Avoid Chinese or special characters in filenames to prevent encoding issues

2. Detect platform and run the upload command:

Windows (PowerShell / CMD)

bash

curl.exe -s -F "file=@<filepath>" -F "expiry=1h" -F "max_visits=0" https://dropage.online/api/upload

macOS / Linux (Bash / Zsh)

bash

curl -s -F "file=@<filepath>" -F "expiry=1h" -F "max_visits=0" https://dropage.online/api/upload

3. Parse the JSON response and report the result to the user.

Response format

Success (HTTP 201):

json

{"success": true, "url": "https://dropage.online/a1b2c3d4e5f6/", "expires_at": "2026-06-29T14:00:00+08:00", "max_visits": 0, "id": "a1b2c3d4e5f6"}

Failure (HTTP 4xx):

json

{"success": false, "error": "reason"}

14-day upload: concrete quota failure examples

Use -i to include the HTTP status and Retry-After header:

bash

# Windows
curl.exe -sS -i -F "file=@page.html" -F "expiry=14d" https://dropage.online/api/upload

# macOS/Linux
curl -sS -i -F "file=@page.html" -F "expiry=14d" https://dropage.online/api/upload

The following examples assume a request at 2026-09-05 23:00 UTC+8. The actual reset is 2026-09-06 00:00 UTC+8, returned as 2026-09-05T16:00:00.000Z (UTC). Retry-After: 3600 is illustrative; use the actual server response. The API's error text is currently Chinese regardless of the page language; interpret the structured fields and explain them in the user's language.

Per-IP daily allowance exhausted: this IP has already successfully uploaded one 14d page today. Another 14d upload is rejected:

http

HTTP/1.1 429 Too Many Requests
Content-Type: application/json; charset=utf-8
Retry-After: 3600

json

{
  "success": false,
  "error": "14 天档当前 IP今日配额已用完（每天 1 个）。请在 UTC+8 次日 00:00 后重试，或选择其他有效期；仍受通用上传频率限制。",
  "code": "DAILY_QUOTA_EXCEEDED",
  "expiry": "14d",
  "scope": "ip",
  "limit": 1,
  "reset_at": "2026-09-05T16:00:00.000Z"
}

Tell the user: “The upload failed because this IP has used today's one-upload allowance for the 14-day tier. It resets at 2026-09-06 00:00 UTC+8. No new page was created.”

Site-wide daily allowance exhausted: ten 14d uploads have already succeeded today. An additional request from an IP that has not used its own allowance is rejected:

http

HTTP/1.1 429 Too Many Requests
Content-Type: application/json; charset=utf-8
Retry-After: 3600

json

{
  "success": false,
  "error": "14 天档全站今日配额已用完（每天 10 个）。请在 UTC+8 次日 00:00 后重试，或选择其他有效期；仍受通用上传频率限制。",
  "code": "DAILY_QUOTA_EXCEEDED",
  "expiry": "14d",
  "scope": "global",
  "limit": 10,
  "reset_at": "2026-09-05T16:00:00.000Z"
}

Tell the user: “The upload failed because today's site-wide allowance of ten 14-day pages is exhausted. It resets at 2026-09-06 00:00 UTC+8. No new page was created.”

When both daily limits are exhausted, the API reports scope: "ip" first. The general upload rate limiter runs before daily quota checks and can instead return a generic HTTP 429 without DAILY_QUOTA_EXCEEDED; do not assume every 429 is a daily quota error. A rejected upload returns no public URL, creates no upload record, and consumes no additional daily quota. Do not loop retries or silently change expiry. Wait until the reported reset, or let the user choose another tier.

Report to the user

On success, show:

• The public URL
• Expiration time (per the chosen expiry; default 1 hour from upload)
• Visit limit, if any (disabled after 1 or 10 visits)

On failure, show the error reason. For HTTP 429, inspect code and Retry-After. When code is DAILY_QUOTA_EXCEEDED, report the affected expiry, scope (ip or global), limit, and reset_at to the user. Wait until reset, or let the user choose another suitable tier; do not immediately retry.

Limits

• File types: .html, .zip, .tar, .tar.gz, .tgz only
• Max size: 30 MB
• Archive must contain index.html at the root level
• Max 50 files per archive
• Expiry: 1h (default) / 6h / 24h / 7d / 14d
• Max visits: 0 (unlimited, default) / 1 / 10
• Storage retention: uploaded files are physically deleted 30 days after upload; database metadata is retained. Expiry and visit limits still disable public access earlier.
• Access log retention: 30 days
• Rate limit: 3 per minute, 30 per half hour, 45 per hour per IP

Examples

Deploy a single HTML file (defaults)

bash

# Windows
curl.exe -s -F "file=@index.html" https://dropage.online/api/upload

# macOS/Linux
curl -s -F "file=@index.html" https://dropage.online/api/upload

Deploy a zip archive

bash

# Windows
curl.exe -s -F "file=@site.zip" https://dropage.online/api/upload

# macOS/Linux
curl -s -F "file=@site.zip" https://dropage.online/api/upload

Deploy a tar.gz archive (Linux/macOS common)

bash

# Windows
curl.exe -s -F "file=@site.tar.gz" https://dropage.online/api/upload

# macOS/Linux
curl -s -F "file=@site.tar.gz" https://dropage.online/api/upload

Deploy with 1-hour expiry, disabled after 1 visit

bash

# Windows
curl.exe -s -F "file=@index.html" -F "expiry=1h" -F "max_visits=1" https://dropage.online/api/upload

# macOS/Linux
curl -s -F "file=@index.html" -F "expiry=1h" -F "max_visits=1" https://dropage.online/api/upload

Deploy with long expiry (daily quotas apply)

7d: 50 successful uploads/day site-wide, 5/IP/day. 14d: 10 site-wide, 1/IP/day. Reset: 00:00 UTC+8.

bash

# Windows: 7 days
curl.exe -s -F "file=@index.html" -F "expiry=7d" https://dropage.online/api/upload
# macOS/Linux: 14 days
curl -s -F "file=@index.html" -F "expiry=14d" https://dropage.online/api/upload

For 6-hour or 24-hour expiry, use -F "expiry=6h" or -F "expiry=24h" respectively.

Deploy from specific path

bash

# Windows
curl.exe -s -F "file=@C:\Users\me\project\index.html" https://dropage.online/api/upload

# macOS/Linux
curl -s -F "file=@/home/user/project/index.html" https://dropage.online/api/upload

Troubleshooting

"file not found" error

• Verify the file path is correct and the file exists
• On Windows PowerShell, use backward slashes or absolute paths: @C:\path\to\file.html
• On macOS/Linux, use forward slashes: @/home/user/file.html
• If the path contains spaces, quote it: @"C:\path with spaces\file.html" (Windows) or @"/path/with spaces/file.html" (macOS/Linux)

Empty file or wrong file type

• File must be non-empty and have one of these extensions: .html, .zip, .tar, .tar.gz, .tgz (case-insensitive)
• Hidden files (e.g. .html files starting with a dot) are not supported

HTTP 400 (Bad Request) — "不支持的文件格式"

• File extension must be one of: .html, .zip, .tar, .tar.gz, .tgz
• Do not use .htm, .xhtml, .rar, .7z, .tar.bz2, .tar.xz, or other formats

HTTP 400 (Bad Request) — Archive errors

• Archive does not contain index.html at the root level (after top-level directory stripping)
• Archive has more than 50 files
• Archive is corrupted, truncated, or uses unsupported compression
• For tar.gz: ensure the file is a valid gzip-compressed tar (not raw gzip of a single file)

HTTP 413 (Payload Too Large)

• File exceeds the 30 MB limit
• Compress images or assets before deploying
• For large projects, consider splitting into multiple deployments

HTTP 429 (Too Many Requests)

• Rate limit exceeded: 3 per minute, 30 per 30 minutes, 45 per hour per IP
• Wait for the cooldown period before retrying
• Daily quota errors have code: DAILY_QUOTA_EXCEEDED plus expiry, scope, limit, and reset_at. Tell the user whether the site-wide or per-IP allowance is exhausted and show the reset time (00:00 UTC+8). Honor Retry-After; do not loop retries or silently choose a shorter expiry.
• An expired or cleaned page still counts toward the day on which it was successfully uploaded. Failed uploads do not consume daily quota.