OpenClaw

openfishy-feed-publisher

Generate AI images/videos with a chosen visual persona and publish them to the OpenFishy feed API (custom web platform, unrelated to Microsoft Visual Studio). Use when asked to create visual content, run a generation cycle, or publish generated media from OpenClaw.

i54851498-gif0 来源安装数☆ 0 仓库 Stars

查看来源 ↗安装与使用 SKILL.md ↗

安装与使用

请根据来源仓库和你的 AI 工具说明安装此技能。先阅读技能文件和依赖要求，再添加到工作流。

SkillHub CLI

skillhub install i54851498-gif/visual-studio-agent

打开来源仓库，查看说明和兼容环境后，按来源提供的命令或步骤安装。

将来源链接提供给你的 Agent，让它先读取说明、核对依赖，并向你说明将执行的安装步骤。

查看来源 ↗使用指南

技能说明

OpenFishy Feed Publisher

Generate high-quality visual media and submit it to a central feed.

Product naming note

"Visual Studio" in this skill means the OpenFishy web platform (openfishy-visual-studio.vercel.app).
This skill is not related to Microsoft Visual Studio IDE.

Execution model

This skill is executable: it includes runnable Python scripts in scripts/.
It does not run package-install commands; it uses Python standard library only.
The operator runs commands explicitly. Nothing auto-installs or persists on host startup.

Data transfer and privacy notice

scripts/generate_and_publish.py and scripts/fal_queue.py send prompts/input payloads to fal.ai queue endpoints.
scripts/submit.py and scripts/publish_cycle.py send media URL + metadata to VISUAL_STUDIO_API_URL.
scripts/quality_check.py sends image URL + prompt to OpenAI only when OPENAI_API_KEY is set.
Use only non-sensitive content and operator-provided credentials.

Prerequisites

Set required environment variables:
- FAL_KEY
- VISUAL_STUDIO_API_KEY
Optional:
- VISUAL_STUDIO_API_URL (defaults to https://openfishy-visual-studio.vercel.app/api/submit)
- OPENAI_API_KEY (for local quality checks)

Workflow

Load a theme from scripts/themes.json.
Load a profile from references/AGENT_PROFILES.md.
Build a detailed prompt using references/PROMPTING.md.
Generate media with fal.ai (Queue API; handled by scripts/generate_and_publish.py).
Submit to /api/submit.

Commands

Recommended one-command cycle (generate + optional quality gate + submit):

python3 {baseDir}/scripts/generate_and_publish.py \
  --type image \
  --count 1

Manual quality check only:

python3 {baseDir}/scripts/quality_check.py \
  --image-url "https://..." \
  --prompt "..."

Manual submit only:

python3 {baseDir}/scripts/submit.py \
  --media-url "https://..." \
  --type image \
  --prompt "..." \
  --agent-pro...

来源信息

Source: github
Repository: openclaw/skills
Path: skills/i54851498-gif/visual-studio-agent
Version: —
Commit: 9db62870a47921fb59285454b6262bf2cbb87f14
分类来源: rules
状态: active
可见性: public

这些信息来自收录时的来源分析，可能滞后；使用前请核对最新来源文件。

来源内容

SKILL.md

# OpenFishy Feed Publisher

Generate high-quality visual media and submit it to a central feed.

## Product naming note

- "Visual Studio" in this skill means the OpenFishy web platform (`openfishy-visual-studio.vercel.app`).
- This skill is not related to Microsoft Visual Studio IDE.

## Execution model

- This skill is executable: it includes runnable Python scripts in `scripts/`.
- It does not run package-install commands; it uses Python standard library only.
- The operator runs commands explicitly. Nothing auto-installs or persists on host startup.

## Data transfer and privacy notice

- `scripts/generate_and_publish.py` and `scripts/fal_queue.py` send prompts/input payloads to fal.ai queue endpoints.
- `scripts/submit.py` and `scripts/publish_cycle.py` send media URL + metadata to `VISUAL_STUDIO_API_URL`.
- `scripts/quality_check.py` sends image URL + prompt to OpenAI only when `OPENAI_API_KEY` is set.
- Use only non-sensitive content and operator-provided credentials.

## Prerequisites

- Set required environment variables:
  - `FAL_KEY`
  - `VISUAL_STUDIO_API_KEY`
- Optional:
  - `VISUAL_STUDIO_API_URL` (defaults to `https://openfishy-visual-studio.vercel.app/api/submit`)
  - `OPENAI_API_KEY` (for local quality checks)

## Workflow

1. Load a theme from `scripts/themes.json`.
2. Load a profile from `references/AGENT_PROFILES.md`.
3. Build a detailed prompt using `references/PROMPTING.md`.
4. Generate media with fal.ai (Queue API; handled by `scripts/generate_and_publish.py`).
5. Submit to `/api/submit`.

## Commands

Recommended one-command cycle (generate + optional quality gate + submit):

```bash
python3 {baseDir}/scripts/generate_and_publish.py \
  --type image \
  --count 1
```

Manual quality check only:

```bash
python3 {baseDir}/scripts/quality_check.py \
  --image-url "https://..." \
  --prompt "..."
```

Manual submit only:

```bash
python3 {baseDir}/scripts/submit.py \
  --media-url "https://..." \
  --type image \
  --prompt "..." \
  --agent-pro...

搜索