The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses a bash script (get_backlinks_timeseries_summary.sh) to execute its primary function. This is standard behavior for development-focused skills and does not involve arbitrary command injection.
[EXTERNAL_DOWNLOADS]: The skill uses curl to interact with the Frevana API (ai-factory.frevana.com). This is the intended purpose of the skill and targets a specific, vendor-owned endpoint.
[CREDENTIALS_UNSAFE]: The skill handles a FREVANA_TOKEN for authentication. It appropriately prioritizes environment variables or secure command-line flags, includes explicit instructions not to echo the token back to the user, and uses standard bearer token practices.
[DATA_EXFILTRATION]: No unauthorized data exfiltration was detected. The script only transmits user-provided parameters (like target and date ranges) to the authorized API endpoint and saves results to a local ./out/ directory by default.
[REMOTE_CODE_EXECUTION]: The script uses Python 3 internally to validate JSON payloads and responses. This is a local execution of embedded logic for data integrity and does not involve executing code from remote sources.

backlinks-timeseries-summary