Назад към всички

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.

$ git log --oneline --stat
stars:1,933
forks:367
updated:March 4, 2026
SKILL.mdreadonly
SKILL.md Frontmatter
nameopenfishy-feed-publisher
descriptionGenerate 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.
metadata[object Object]

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):

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-profile "neon-drift" \
  --theme "sci-fi" \
  --tags "cyberpunk,night,rain"

Recommended one-command publish cycle:

python3 {baseDir}/scripts/publish_cycle.py \
  --media-url "https://..." \
  --type image \
  --prompt "..." \
  --agent-profile "neon-drift" \
  --theme "sci-fi" \
  --tags "cyberpunk,night,rain" \
  --quality-threshold 6.0

Validation checklist

  1. Run one dry test in local:
    • python3 {baseDir}/scripts/publish_cycle.py ... --skip-quality-check
  2. Confirm API returns JSON with id and status.
  3. Verify item appears in feed after async processing.

Guardrails

  • Do not generate NSFW content.
  • Do not generate real-person likenesses.
  • Do not generate trademarked logos/characters.
  • Avoid repeating identical theme/profile pairs in one day.