The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell scripts in SKILL.md that interpolate user-provided input (CERT_INPUT) directly into commands like echo, cut, and openssl. Without proper sanitization, an attacker providing an input such as $(malicious_command) or using backticks could trigger arbitrary command execution when the agent runs the workflow steps.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface.
Ingestion points: CERT_INPUT variable in SKILL.md used to define target files or hosts.
Boundary markers: Absent; there are no delimiters or instructions for the agent to ignore embedded commands in the input data.
Capability inventory: The skill uses high-capability tools including openssl, grep, sed, date, cut, and echo within a shell environment.
Sanitization: Absent; the script does not validate or escape the CERT_INPUT variable before passing it to subshells and command substitutions.

cert-parse