The Agent Skills Directory

[DATA_EXPOSURE_AND_EXFILTRATION]: The skill securely handles authentication by retrieving the ATTIO_API_KEY through the platform's get-config utility with the includeSecrets=true flag. All network requests are directed to the official Attio API at api.attio.com. No evidence of unauthorized data transmission or credential leakage was found.
[COMMAND_EXECUTION]: The skill uses standard shell commands (curl, jq, sleep) to perform CRM operations. These commands are structured as templates and do not involve the execution of untrusted external scripts or arbitrary user input without context.
[INDIRECT_PROMPT_INJECTION]: As the skill ingests data from an external CRM source (Attio), it theoretically possesses an attack surface for indirect prompt injection. However, the ingested data is processed using structured tools like jq and the risk is inherent to the primary function of a CRM integration tool, presenting no immediate security threat in this context.

attio-interaction