The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes curl for making HTTP requests to the Close CRM API and jq for processing JSON data in the terminal. These are standard tools for API interaction.
[DATA_EXFILTRATION]: The skill transmits data to https://api.close.com. This is the official and documented endpoint for the Close CRM service, making the network operations consistent with the skill's stated purpose.
[CREDENTIALS_UNSAFE]: Authentication is handled via the $CLOSE_TOKEN environment variable. This is a secure practice that avoids hardcoding sensitive API keys directly within the skill instructions.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external API responses (Close CRM). While this creates a theoretical surface for indirect prompt injection if an attacker controls CRM data, the skill does not grant excessive privileges or perform dangerous operations with the ingested data, presenting a minimal risk profile.

close