shellbot-creative

Opinionated creative production system for image/video generation, image editing, motion scenes, voiceovers, music, and Remotion assembly. Combines Freepik,...

Security Scan

The skill's code, scripts, and runtime instructions align with its stated purpose as a multi-provider creative production orchestrator; requested credentials and network calls are proportionate to that purpose.

Purpose & Capability

Name/description promise (multi-provider creative pipeline using Freepik, fal.ai, Nano Banana 2, and Remotion) matches the contents: orchestration scripts, Remotion React templates, and numerous curl/python commands targeting those providers. The declared primary credential (FREEPIK_API_KEY) and providerEnv references to FAL_KEY/INFERENCE_API_KEY are appropriate. Minor metadata inconsistency: registry 'Required env vars: none' conflicts with the skill's primaryEnv and the SKILL.md checks which expect FREEPIK_API_KEY (and optionally FAL_KEY / INFERENCE_API_KEY).

Instruction Scope

SKILL.md and scripts instruct running local Python scripts, creating project folders, using Remotion (node/npx), and calling provider APIs via curl/infsh. All referenced network endpoints are provider endpoints (api.freepik.com, queue.fal.run / api.fal.ai, infsh/inference.sh); instructions do not reference reading unrelated system files or exfiltrating files to unknown hosts. The runtime has permission to read/write local project files (assets, manifests) as expected.

Install Mechanism

No automated install spec in the registry (instruction-only), which minimizes automatic code fetching. The repo includes packaging/install helper scripts (scripts/install_skill.sh, package_skill.sh) — their contents were not fully inspected here; they are typical but should be reviewed before executing. Visible files do not download arbitrary archives from unknown hosts.

Credentials

Requested credentials are proportional to the multi-provider workflow: FREEPIK_API_KEY is the primary credential and FAL_KEY / INFERENCE_API_KEY are optional fallbacks. No unrelated secrets or broad system credentials are requested. Reminder: API keys will be transmitted to the provider endpoints invoked by the scripts (normal for this skill).

Persistence & Privilege

The skill is not always-enabled and uses default agent autonomy settings. It does not request elevated system privileges or indicate modification of other skills or global agent configuration. It writes artifacts into local project folders (assets/, scenes/, audio/, manifests/) which is consistent with its purpose.

Assessment

This skill appears coherent for orchestrating multi-provider creative workflows, but review these before installing or running: 1) Supply credentials only for providers you intend to use (FREEPIK_API_KEY is required for Freepik flows); understand that those keys will be sent to the provider APIs and may incur usage/billing. 2) Inspect scripts/install_skill.sh and package_skill.sh if you plan to run them — don't execute install scripts from unknown packages without reading them. 3) Review any truncated/omitted files for unexpected network endpoints or obfuscated code. 4) If you want to limit blast radius, create provider API keys with the narrowest scopes possible and run the skill in an isolated project folder. 5) If you need higher assurance, run a dry-run (the repo includes dry-run manifests) and review the generated shell plans before executing any network calls.

Create high-end creative outputs by chaining providers intentionally:

1. Generate reusable assets (Nano Banana 2 / Freepik / fal)
2. Generate scene motion (Freepik Kling / fal video models)
3. Add voice and music (Freepik ElevenLabs voiceover + music or fal equivalents)
4. Assemble and polish in Remotion

Use this skill as a production orchestrator, not as isolated single calls.

Arguments

• Command: $0 (plan | asset | edit | video | voice | music | remotion | pipeline | status | sample)
• Arg 1: $1 (provider, model, or workflow type)
• Arg 2+: $2, $3, etc.
• All args: $ARGUMENTS

Authentication and provider checks

Support any of these providers:

• FREEPIK_API_KEY for Freepik API
• FAL_KEY for fal.ai API
• INFERENCE_API_KEY (or infsh login) for Nano Banana 2 via inference.sh

Before any generation call:

[ -n "$FREEPIK_API_KEY" ] && echo "Freepik ready"
[ -n "$FAL_KEY" ] && echo "fal ready"
command -v infsh >/dev/null && echo "infsh available"

If a requested provider is not authenticated, route to another available provider and clearly explain the fallback.

Provider routing (opinionated defaults)

• Asset ideation / strong multi-image consistency: Nano Banana 2
• Photoreal product hero, posters, typography, upscale/edit suite: Freepik
• Fast model exploration or niche fal endpoint needs: fal.ai
• Final timeline composition, transitions, captions, audio mix: Remotion

Read references/provider-matrix.md for exact model choices.

Command behavior

$0 = plan

Turn a user brief into a production storyboard and shot list.

python3 scripts/creative_brief_to_storyboard.py \
  --brief "Launch video for a new fitness app aimed at busy professionals" \
  --format product-marketing \
  --duration 45 \
  --aspect-ratio 16:9 \
  --out storyboard.json

Then create a provider/model routing recommendation:

python3 scripts/creative_provider_router.py \
  --goal full-video \
  --priority quality \
  --needs-consistency true \
  --needs-typography true

$0 = asset

Generate still assets for scenes.

• Nano Banana 2 default (consistency + editing):

infsh app run google/gemini-3-1-flash-image-preview --input '{
  "prompt": "Premium SaaS dashboard on a laptop, soft studio light",
  "aspect_ratio": "16:9",
  "num_images": 4,
  "resolution": "2K"
}'

• Freepik high-fidelity product image (Mystic):

curl -s -X POST "https://api.freepik.com/v1/ai/mystic" \
  -H "x-freepik-api-key: $FREEPIK_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"Studio product shot of matte black earbuds on reflective surface","resolution":"2k","styling":{"style":"photo"}}'

• fal fast concepting (Flux):

curl -s -X POST "https://queue.fal.run/fal-ai/flux-2" \
  -H "Authorization: Key $FAL_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"Cyberpunk storefront, rainy night","image_size":"landscape_16_9"}'

$0 = edit

Perform image editing/upscaling/background work before animation.

• Prefer Freepik edit endpoints for deterministic production flows.
• Prefer Nano Banana 2 when iterative instruction-based edits are needed with multiple references.
• Use Freepik upscalers for final delivery quality.

See references/workflow-recipes.md for edit chains.

$0 = video

Generate motion scenes from prompts or keyframes.

• Freepik default for premium multi-shot video: kling-v3-omni-pro
• fal default when Freepik key is unavailable: fal-ai/kling-video/v2/image-to-video

Always save scene outputs in a structured project folder:

mkdir -p ./creative-output/{assets,scenes,audio,final,manifests}

$0 = voice

Generate narration.

• Default: Freepik ElevenLabs voiceover endpoint
• Fallback: fal text-to-speech model if user requests fal-only flow

Use concise script-first narration lines generated from storyboard scenes.

$0 = music

Generate background music and keep loudness headroom for voiceover.

• Default: Freepik music-generation
• Target: deliver 10-240s cue matching storyboard duration

$0 = remotion

Assemble final video with composited assets, generated clips, captions, and audio mix.

npx create-video@latest creative-video
cd creative-video

Generate a Remotion manifest from the storyboard:

python3 ../scripts/remotion_manifest_from_storyboard.py \
  --storyboard ../storyboard.json \
  --fps 30 \
  --voiceover-url "https://example.com/voice.mp3" \
  --music-url "https://example.com/music.mp3" \
  --out ./src/creative-manifest.json

Then implement composition using assets/remotion/ProductMarketingTemplate.tsx as a baseline.

$0 = pipeline

Run end-to-end sequence:

1. plan from brief
2. asset generation per scene
3. edit and upscale selected assets
4. video generation for hero scenes
5. voice + music
6. remotion assembly + render

If the user asks for a product marketing or explainer video, this is the default command path.

$0 = status

Check async task status for Freepik/fal requests and report:

• current state
• queue position if available
• failed-step diagnosis
• next action

$0 = sample

Load ready-to-run workflows from references/samples.md.

Quality bar for "outstanding" creative output

Always enforce:

• Clear narrative arc: hook -> value -> proof -> CTA
• Visual consistency across scenes (same product, palette, style)
• Voice/music balance (voice intelligibility first)
• Captions and on-screen text optimized for mobile safe zones
• Brand-safe outputs and explicit user approval before final render

Production workflow policy

• Prefer 16:9 for explainers and product demos unless user asks otherwise
• Prefer 9:16 for shorts/reels/tiktok style cuts
• Generate 2-4 variations of key hero scenes, then converge
• Keep source assets reusable (clean backgrounds, layered variants when possible)
• For revisions, edit prompts/scene timing first before regenerating everything

Local resources included in this skill

• scripts/creative_brief_to_storyboard.py - brief -> timed scene plan
• scripts/creative_provider_router.py - requirement -> provider/model route
• scripts/remotion_manifest_from_storyboard.py - storyboard -> Remotion-ready manifest
• scripts/run_full_dry_run.py - full local dry run that outputs storyboard, routes, manifest, and runnable API plan
• scripts/package_skill.sh - build a distributable .tar.gz of the skill
• scripts/install_skill.sh - install the skill folder into an OpenClaw skills directory
• references/creative-guidelines.md - durable standards for creative quality
• references/provider-matrix.md - provider/model decision matrix
• references/workflow-recipes.md - end-to-end recipes for common outcomes
• references/samples.md - concrete command samples
• references/remotion-playbook.md - Remotion-specific finishing guidance
• assets/remotion/ProductMarketingTemplate.tsx - starter Remotion composition
• assets/remotion/Root.tsx - composition registration sample

References

• OpenClaw skills format and frontmatter: https://docs.openclaw.ai/tools/skills
• OpenClaw slash command behavior ($0, $1, $ARGUMENTS): https://docs.openclaw.ai/tools/slash-commands
• OpenClaw metadata config (metadata.openclaw): https://docs.openclaw.ai/tools/skills-config
• OpenClaw creating skills guide: https://docs.openclaw.ai/tools/creating-skills