The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data through the $ARGUMENTS variable to generate long-form implementation plans and goal-setting documents. This creates a vulnerability where instructions embedded within a user's objective could influence the agent's behavior during the planning phase.
Ingestion points: User input provided as the objective ($ARGUMENTS) in SKILL.md.
Boundary markers: None; the skill does not use specific delimiters or instructions to the agent to ignore potentially malicious commands within the user objective.
Capability inventory: The skill has the capability to write multiple markdown files to the project directory and modify version control configuration files (.gitignore).
Sanitization: There is no evidence of validation or sanitization of the user input before it is interpolated into the generated artifacts.
[COMMAND_EXECUTION]: The skill instructs the agent to perform file system and configuration tasks, specifically creating new directories for plans and appending lines to the .gitignore file to prevent local artifacts from being committed to the repository. While these are administrative tasks, they involve direct modification of project configuration based on skill logic.

plan-goal