The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes multiple shell commands including openssl, cp, sed, grep, and date. These are used for the intended purpose of certificate analysis and verification, but provide a significant execution capability surface.
[INDIRECT_PROMPT_INJECTION]: The skill processes external hostnames and file paths which may be provided by an agent processing untrusted data.
Ingestion points: The CERT_INPUT and CA_BUNDLE variables in SKILL.md are the primary entry points for external data.
Boundary markers: Absent. The skill does not use specific delimiters or instructions to ignore potential injections within the provided input strings.
Capability inventory: The skill possesses capabilities for network access via openssl s_client, file reading via openssl x509 -in and [ -f ], and file writing to the /tmp directory via cp and openssl -out.
Sanitization: The script employs double-quoting (e.g., "$HOST", "$WORK_CERT") which effectively mitigates direct shell command injection. However, it lacks validation for hostnames or file paths to prevent access to unauthorized local files or unintended network destinations.

cert-parse