The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash(cmd:*) tool to run internal scripts, specifically check_cert_expiry.sh, renew_cert.sh, and list_certs.sh. These scripts are designed to perform administrative tasks such as querying certificate status and modifying system configurations for renewals. This capability is expected for the skill's primary function of system-level certificate management.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data that could be controlled by an external attacker, such as fields within an SSL/TLS certificate retrieved from a domain.
Ingestion points: Certificate metadata (issuer, subject, SANs, expiry) retrieved via list_certs.sh or check_cert_expiry.sh and user-provided domain strings (SKILL.md).
Boundary markers: No specific boundary markers or "ignore" instructions for processed data are defined in the instructions (SKILL.md).
Capability inventory: The agent has access to Bash(cmd:*), Read, Grep, and Glob tools (SKILL.md).
Sanitization: The instructions do not specify any validation or sanitization procedures for certificate content before it is parsed by the agent or presented to the user.
[EXTERNAL_DOWNLOADS]: The certificate renewal logic (renew_cert.sh) interacts with Let's Encrypt and other ACME providers. Let's Encrypt is a well-known and trusted certificate authority, and this network interaction is a legitimate component of the SSL lifecycle management provided by the skill.

managing-ssltls-certificates