{"id":"zerolu-ultimate-ai-media-generator-skill","name":"Ultimate-AI-Media-Generator-Skill","af_score":49.2,"security_score":48.0,"reliability_score":27.5,"what_it_does":"Provides an open-source “skill” for AI coding/agent platforms that can generate images and videos using CyberBara-backed models (e.g., Nano Banana, Sora, Seedance, Kling/Ve o), including helper commands for credit quotation/balance checks and task waiting to retrieve final media URLs.","best_when":"You want an agent skill that can reliably call image/video generation models with prebuilt prompts and simple operational steps (quote, generate, wait, retrieve URLs) across common agent platforms.","avoid_when":"You require strong guarantees about error codes, rate-limit headers, idempotency semantics, and detailed reliability/SLA documentation from the package itself.","last_evaluated":"2026-03-30T15:30:40.295212+00:00","has_mcp":false,"has_api":false,"auth_methods":["API key configured via scripts/cyberbara_api.py setup-api-key and stored locally at ~/.config/cyberbara/api_key"],"has_free_tier":false,"known_gotchas":["Skill relies on an external provider (CyberBara) and model availability; fallback behavior is mentioned only for one example (seedance-2.0-pro -> sora-2).","README suggests a local file for API key storage, which may impact sandboxed agents or ephemeral environments.","No explicit guidance is given for idempotent retries; repeated generation requests may incur additional credits if not handled carefully by the agent."],"error_quality":0.0}