AI News Tracker

This skill is coherent with its purpose, but review a few practical privacy and operational points before installing: - Third-party fetch/extract: it uses Jina Reader (https://r.jina.ai/...) to convert pages to plain text. That means article content is proxied to Jina's service — if you need to avoid sending page contents to external services, do not enable this skill or request an alternative extractor. - Search/summarize calls: batch_web_search and summarize steps will send query text and retrieved content to the platform's search/summarization services (and likely the LLM). Expect external network calls and potential data exposure of fetched content. - Workspace file access: the skill reads and writes `[WORKSPACE]/memory/ai-news-draft.md` and expects to read a Feishu user ID from workspace memory. Make sure no sensitive secrets are stored in that workspace path and that the Feishu target is correct. - Feishu sending: the skill assumes an existing Feishu connector/credentials managed by the platform; verify what credentials/scopes the platform has for Feishu and whether you are comfortable with automated pushes. - Portability/detail: the SKILL.md uses GNU date syntax (date -d) which may not work on all hosts; this is an operational (not security) detail to check. If these exposures are acceptable for your environment (i.e., you allow content to be proxied to the indicated services and you trust the platform's Feishu connector), the skill appears internally consistent and appropriate for its stated purpose.