The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions direct the agent to pass user-provided input directly into a Bash command shell.
Evidence: The instruction taskmd next $ARGUMENTS in SKILL.md interpolates untrusted user input into a command-line execution string without specifying validation or escaping.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted data from multiple sources and provides it to the agent context.
Ingestion points: User-supplied $ARGUMENTS and the content of task files accessed via the Read tool.
Boundary markers: Absent; there are no instructions to the agent to ignore or delimit embedded instructions within the task files or arguments.
Capability inventory: The skill utilizes the Bash tool for command execution and the Read tool for file system access.
Sanitization: Absent; the skill does not define any filtering or escaping mechanisms for the processed data.

next-task