Podcast Agent

Search articles on any topic, generate a two-host dialogue script, and synthesize podcast audio via TTS. Turn long reads into listenable content.

This skill appears to implement the podcast workflow it claims, but the metadata wrongly lists 'edge-tts' as a required environment variable even though the code imports the edge_tts Python package; this is likely a packaging/metadata error. Before installing or running: (1) treat it as code that will fetch arbitrary URLs and call network TTS services — avoid passing private/internal URLs or secrets; (2) inspect and, if necessary, run the script in a sandboxed environment; (3) install edge-tts from PyPI only if you trust that package (pip install edge-tts is suggested); (4) you don't need to set credentials for this skill as distributed, but verify whether you want to provide any API keys or private endpoints — the skill will not ask for them. If you plan to deploy widely, ask the author to fix the metadata (remove or correct the 'edge-tts' env var) so it's clear what secrets, if any, are required.

Type: OpenClaw Skill Name: podcast-agent Version: 1.0.2 The podcast-agent skill is a legitimate tool designed to convert web articles into audio podcasts using edge-tts. The core logic in scripts/podcast_gen.py includes proactive security measures, such as a robust SSRF validation function that blocks access to internal hostnames and private IP ranges. The script handles subprocess execution safely without shell=True, and the instructions in SKILL.md are consistent with the stated purpose without any evidence of prompt injection or malicious intent.