The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses curl to make HTTP requests to the 6551.io API. This is the primary intended function of the skill for data retrieval.
[CREDENTIALS_UNSAFE]: The skill requires an OPENNEWS_TOKEN environment variable for authentication. It provides clear instructions for users to obtain their own token and does not contain any hardcoded secrets.
[EXTERNAL_DOWNLOADS]: The skill makes network requests to ai.6551.io. These are legitimate API calls to the service provider described in the documentation.
[DATA_EXFILTRATION]: No patterns of unauthorized data access or transmission were detected. The data flows are limited to sending search parameters to the official API and receiving news content.
[INDIRECT_PROMPT_INJECTION]: The skill processes external news data (titles, summaries). While this presents a theoretical injection surface common to all news-reading tools, it is used for information display and does not have high-privilege capabilities that would lead to escalation.

opennews