The Agent Skills Directory

[DATA_EXFILTRATION]: The skill instructs the agent to read sensitive local files, including .env, config, and environment files, to collect evidence during the troubleshooting process. While the skill includes instructions to mask secrets, accessing these files directly creates a risk of exposing credentials or sensitive configuration data within the session context.
[COMMAND_EXECUTION]: The instructions require the agent to write detailed debug logs to a local file named antom_debug.log in the project root. This log is intended to contain complete API requests and responses, which may inadvertently include sensitive data on the local filesystem if masking is not applied perfectly to all fields.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data by reading local log files and user-provided evidence during troubleshooting. This creates an attack surface where malicious content embedded in logs could attempt to override agent behavior. The risk is mitigated by explicit instructions to mask secrets and follow a specific diagnosis ledger.
[EXTERNAL_DOWNLOADS]: The skill fetches documentation and guides from official Antom domains including cdn.marmot-cloud.com and dashboard.antom.com. These are recognized as legitimate vendor resources for the Antom payment service and are documented here as part of the skill's normal operation.

antom-integration