An automated script generator for creating viral videos.

Story generation pipeline skill. Supports multi-episode continuous generation, graph management, AI quality check + human confirmation dual control mechanism...

This skill largely matches its stated purpose (local story pipeline) but has a few red flags you should consider before installing or running it: 1) Documentation vs. code mismatches: SKILL.md talks about a remote Graph API and enforces 'max 3 retries' for AI review, but the code stores graphs locally and allows unlimited AI retries. That could be a stale doc or an incomplete implementation — ask the author or inspect runtime behavior. 2) Data persistence: the skill saves full episode content and pipeline state under data/, which can include any user-provided or model-generated sensitive text; treat those files as potentially sensitive and consider where you run the skill and applying access controls or encryption. 3) Resource exhaustion: unlimited AI retry loops (as implemented) can lead to repeated generation attempts and excessive compute or cost if an LLM is invoked externally — consider adding a retry cap or safeguards. 4) Network behavior: the code as provided makes no outbound network calls, but the docs hint at a remote API that is not implemented; before trusting the skill, run it in a sandboxed environment and monitor for unexpected network activity. 5) Origin and maintenance: there is no homepage and source owner is unknown; if you plan to use this in production, request provenance, tests, and an explicit maintenance plan. Recommended steps: run the code in an isolated environment, inspect and (if needed) modify pipeline.py to enforce retry limits and to sanitize or limit stored content, review and lock file permissions for data/, and confirm whether any external endpoints should be used or are missing. If you need higher assurance, ask the publisher for clarification about the documented Graph API and the intended retry policy.

Type: OpenClaw Skill Name: story-generator Version: 1.0.1 The skill bundle implements a structured story generation pipeline featuring AI-driven quality scoring and human-in-the-loop confirmation. The code (pipeline.py, graph_manager.py) manages story states and relationship graphs using local JSON storage in a dedicated data directory. No evidence of data exfiltration, unauthorized network access, or malicious prompt injection was found; the logic is entirely consistent with the stated purpose of creative writing automation.