Post Content Moderation

Review, rewrite, and moderate user-generated posts across title, body text, images, and videos to block ads and contact information while allowing configurab...

This skill's docs describe a networked moderation integration that will commonly send post text and media URLs to external model APIs and callback endpoints, but the package metadata is inconsistent: it declares no environment secrets and only lists python3 as a required binary while the docs reference bundled PHP demo scripts (which are not present). Before installing or using in production: 1) Ask the publisher to explain why python3 is required and to provide the missing demo scripts or remove those references; 2) Require a clear list of environment variables (API keys, allowed_hosts, MODERATION_DRY_RUN) and verify they are declared; 3) Run in dry-run mode and enable strict allowed_hosts to prevent unintended data exfiltration; 4) Do not provide real API keys or sensitive data until you can audit the actual code that will send data externally; 5) If you need a purely local moderation engine, do not use this skill as-is — it’s designed around networked model/callback workflows. If the publisher cannot clarify or provide the expected code and manifest fixes, treat the package as untrusted.

Type: OpenClaw Skill Name: post-content-moderation Version: 1.3.0 The bundle provides a comprehensive framework for AI-driven content moderation of posts and comments, focusing on blocking advertisements and contact information. The instructions in SKILL.md and the supporting documentation in the references/ directory are well-structured and transparent about the skill's networked capabilities, such as sending data to external APIs for moderation and using callbacks for results. Notably, the documentation includes explicit security warnings, recommends using environment variables for secrets, and describes a host allowlist feature to mitigate unauthorized data exfiltration. There is no evidence of malicious intent, obfuscation, or unauthorized access to sensitive local data.