The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses local shell scripts to manage API requests. This is a standard architectural choice for CLI-based agents and is implemented using defensive bash flags (-euo pipefail) to ensure robust execution.
[CREDENTIALS_UNSAFE]: The skill correctly manages the FREVANA_TOKEN by reading it from the environment or secure interactive prompts. It includes explicit instructions not to echo the token back to the user or include it in standard output.
[EXTERNAL_DOWNLOADS]: The skill communicates exclusively with the established vendor endpoint https://ai-factory.frevana.com to fetch backlink data. This is consistent with the skill's primary function and targets the author's own infrastructure.
[DATA_EXFILTRATION]: No unauthorized data exfiltration was detected. Network operations are limited to the intended API provider, and the payload is strictly filtered to only include supported fields. Input data like targets and tags are passed directly to the provider for processing.
[DYNAMIC_EXECUTION]: The script uses Python for JSON payload construction and response validation. The code is embedded as a static heredoc string within the shell script, preventing runtime modification or injection of untrusted external scripts.

backlinks-bulk-backlinks