← 返回 Skills 市场
zoucdr

Carpet Wash Video

作者 parallel world · GitHub ↗ · v0.1.0 · MIT-0
cross-platform ⚠ suspicious
110
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install carpet-wash-video
功能描述
Generate satisfying vertical carpet deep-clean shorts (WeryAI): text-to-video or dirty rug photo to rinse, grime runoff, and fiber revival. Use when you need...
使用说明 (SKILL.md)

Carpet wash & restore videos

For cleaning / satisfying creators: pressure rinse, brush agitation, dirty water running out, fibers standing up again—strong before/after and ASMR wash sounds. One prompt or one dirty carpet photo.

Dependencies: scripts/video_gen.js + WERYAI_API_KEY + Node.js 18+.

Prerequisites

  • WERYAI_API_KEY must be set.
  • Node.js 18+; images must be https URLs.

Security, secrets, and API hosts

  • WERYAI_API_KEY: Treat as a secret. Only configure it if you trust this skill's source; it is listed in OpenClaw metadata as requires.env / primaryEnv so installers know it is mandatory at runtime (never commit it inside the skill package).
  • Optional URL overrides (WERYAI_BASE_URL, WERYAI_MODELS_BASE_URL): video_gen.js defaults to https://api.weryai.com and https://api-growth-agent.weryai.com. Overrides are intended for testing or approved alternate endpoints. If these variables are set in your environment, confirm they point to hosts you trust—otherwise prompts, images, and your bearer token could be sent elsewhere.
  • Higher assurance: Run generation in a short-lived or isolated environment (separate account or container), and review scripts/video_gen.js (HTTPS submit + poll loop) before production use.

Prompt expansion (mandatory)

video_gen.js does not expand prompts. Before every wait --json, turn the user's short or vague brief into a full English production prompt.

When: The user gives only keywords, one line, or loose intent—or asks for richer video language. Exception: They paste a finished long prompt within the model's prompt_length_limit and ask you not to rewrite; still show the full text in the confirmation table.

Always add (video language): shot scale and angle; camera move or lock-off; light quality and motivation; subject action paced to duration; one clear payoff for this niche; state 9:16 vertical when this skill defaults to vertical.

Length: Obey prompt_length_limit for the chosen model_key when this doc lists it; trim filler adjectives before removing core action, lens, or light clauses.

Confirmation: The pre-submit table must include the full expanded prompt (never a one-line summary). Wait for confirm or edits.

Niche checklist

  • Wash physics: water/foam migration, dirty runoff, fiber lift, before/after color stripe or patch.
  • Camera: top-down or low side angle; slow motion on squeeze or extractor pass; ASMR-forward if audio on.
  • Stains: tie language to user rug type (mud, pet, grey traffic) and method (pressure, brush, steam).

### Example prompts at the top of this file are short triggers only—always expand from the user's actual request.

Workflow

  1. Confirm scenario (text-to-video and/or image-to-video as documented).
  2. Collect the user's brief, optional https image URL, tier (best / good / fast) or explicit model.
  3. Expand prompt (mandatory): Unless the user supplied a finished long prompt and asked not to rewrite, expand the brief with full shot, light, wash physics, and audio cues per ## Prompt expansion (mandatory). Do not submit a one-liner.
  4. Check the expanded prompt against prompt_length_limit if listed for the model; trim if needed.
  5. Verify duration, aspect_ratio, generate_audio, and other fields against this doc.
  6. Show the confirmation table with the full expanded prompt; wait for confirm or edits.
  7. After confirmation, run node {baseDir}/scripts/video_gen.js wait --json '...' with the expanded prompt.
  8. Return playable URL(s) or clear error guidance.

CLI reference

node {baseDir}/scripts/video_gen.js wait --json '{"model":"…","prompt":"…","duration":5,"aspect_ratio":"9:16"}'
node {baseDir}/scripts/video_gen.js wait --json '…' --dry-run
node {baseDir}/scripts/video_gen.js status --task-id \x3Cid>

Definition of done

Playable URL(s) or clear failure; parameters within model limits. The submitted prompt must be the expanded production prompt unless the user explicitly supplied a finished long prompt and asked not to rewrite it.

Boundaries (out of scope)

  • No legal/commercial guarantees; no offline editing stacks; {baseDir} only—no absolute paths.

Example prompts

  • Filthy living-room rug, pressure washer leaves a clean stripe, vertical satisfying
  • From this moldy carpet photo—rinse and dark water pouring off
  • OCD clean macro: pile goes grey to bright, foam and squeeze moment
  • Satisfying carpet deep clean 9:16, dirty-to-clean reveal line

Default parameters

Field Value
Model KLING_V3_0_PRO
Aspect 9:16
Duration 5 (short hook)
Audio On (brush, water, squeeze ASMR)
Look Top-down close, soft light, extreme before/after color, slow grime flow, minimal background

API validity (KLING_V3_0_PRO): Text: duration 5 / 10 / 15, aspect_ratio 9:16, 1:1, 16:9; image: aspect_ratio 9:16, 16:9, 1:1; no resolution. VEO fast: VEO_3_1_FAST / CHATBOT_VEO_3_1_FAST, duration 8, aspect_ratio 9:16 or 16:9. Other keys: follow tables here.


Text-to-video: wash process

User provides: rug type (short pile / shag / weave / vintage / pet mat), stain type (overall grey / embedded fur / mud spot / drink / years of dust), optional method (pressure / brush / steam / extractor).

Flow: collect → build English prompt (runoff, color return, fiber lift) → run:

node {baseDir}/scripts/video_gen.js wait --json '{"model":"KLING_V3_0_PRO","prompt":"(English prompt)","aspect_ratio":"9:16","duration":5,"generate_audio":true}'

Parameters: model KLING_V3_0_PRO, 9:16, 5, generate_audio true.

Expanded prompt: Build per ## Prompt expansion (mandatory) from the user's rug/stain brief; do not paste fixed samples.

Expected: Visible dirty water, strong color bounce-back, ASMR-friendly audio.


Image-to-video: clean the photo rug

User provides: https image URL + effect (pressure / brush / extract / steam).

Flow: validate URL → prompt matched to stains in image →

node {baseDir}/scripts/video_gen.js wait --json '{"model":"KLING_V3_0_PRO","prompt":"(English)","image":"(URL)","aspect_ratio":"9:16","duration":5,"generate_audio":true}'

Expected: Grime matches photo; true color emerges; pattern continuity.


Prompt blocks

Grime: thick dark water flows outward, murky brown runoff cascades, grime releases in satisfying rivulets

Color return: vivid original color emerges beneath, saturated hues pop, before-after color contrast in single frame

Pile: fibers lift and separate, pile stands upright after cleaning, fluffy texture restored

Sound: ASMR scrubbing, pressure washer hiss, wet squeegee drag

Upload local shots to a public host first; API needs reachable HTTPS.

安全使用建议
This package appears to do exactly what it says: call WeryAI to generate short carpet-cleaning videos. Before enabling, verify you trust the source and do not hardcode your WERYAI_API_KEY into the skill. If you plan to use alternative endpoint env vars (WERYAI_BASE_URL or WERYAI_MODELS_BASE_URL), confirm they point to trusted hosts — changing them can redirect your API key and data. Consider running the script in an isolated container or ephemeral account for higher assurance, and inspect scripts/video_gen.js yourself (it is small and network-only) if you have any doubt. Finally, follow the SKILL.md confirmation workflow so prompts, uploaded image URLs, and the expanded production prompt are reviewed by the user before submission.
能力评估
Purpose & Capability
Name/description (carpet-cleaning short videos) match the declared runtime needs: Node.js and a WERYAI_API_KEY. The primary credential and binaries are what you'd expect for a remote video-generation API client.
Instruction Scope
SKILL.md confines runtime actions to: expanding prompts, validating inputs (https image URLs), and calling the WeryAI API via scripts/video_gen.js. It explicitly warns about secrets and endpoint overrides and does not instruct the agent to read unrelated files or other environment secrets.
Install Mechanism
There is no install spec (instruction + included CLI script). No downloads from untrusted URLs or archive extraction. The shipped Node script is self-contained and targets Node 18+ (which provides fetch).
Credentials
The only required environment variable is WERYAI_API_KEY (declared as primary). Optional overrides (WERYAI_BASE_URL, WERYAI_MODELS_BASE_URL) are documented. No unrelated tokens, credentials, or config paths are requested.
Persistence & Privilege
Skill is not 'always' enabled and is user-invocable. It does not request modifications to other skills or system-wide settings and does not persist additional credentials beyond using the provided API key at runtime.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install carpet-wash-video
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /carpet-wash-video 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v0.1.0
Initial release of "carpet-wash-video" for auto-generating vertical, satisfying carpet cleaning shorts. - Generates text-to-video or image-to-video (dirty carpet photos to rinse and reveal) - Designed for strong before/after, ASMR, grime runoff, and fiber revival effects - Requires WERYAI_API_KEY and Node.js 18+ - Prompts are expanded for rich, shot-accurate, and niche-specific video descriptions - Includes workflow, CLI usage, parameter defaults, and robust prompt confirmation steps
元数据
Slug carpet-wash-video
版本 0.1.0
许可证 MIT-0
累计安装 1
当前安装数 0
历史版本数 1
常见问题

Carpet Wash Video 是什么?

Generate satisfying vertical carpet deep-clean shorts (WeryAI): text-to-video or dirty rug photo to rinse, grime runoff, and fiber revival. Use when you need... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 110 次。

如何安装 Carpet Wash Video?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install carpet-wash-video」即可一键安装,无需额外配置。

Carpet Wash Video 是免费的吗?

是的,Carpet Wash Video 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Carpet Wash Video 支持哪些平台?

Carpet Wash Video 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Carpet Wash Video?

由 parallel world(@zoucdr)开发并维护,当前版本 v0.1.0。

💬 留言讨论