The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses node to execute local scripts (collect-topic-signals.mjs, check-topic-contract.mjs) and opencli to fetch platform data. These are standard operations for the skill's stated purpose of data collection and contract validation within the user's project directory.
[EXTERNAL_DOWNLOADS]: Fetches trending data from aihot.virxact.com using a standard HTTP request. This is a functional requirement for trend scanning and does not involve executable code or sensitive data exfiltration.
[DATA_EXPOSURE]: The skill reads from local JSON and Markdown files (tracked-creators.json, self-channel.json, Jeffrey个人表达池.md) to establish baseline metrics and candidate signals. Access is restricted to the specific episode and data layer directories required for the workflow.

laohan-xuanti