The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local bash script located at .specify/scripts/bash/create-cleanup.sh. This script is invoked using shell command strings where user input from $ARGUMENTS is directly interpolated. This creates a risk of command or argument injection if the input contains shell metacharacters or crafted flags.
[COMMAND_EXECUTION]: The agent is instructed to perform file system operations including moving, renaming, and creating directories and files (e.g., specs/bugfix/001-*). While the skill claims to restrict these actions to the specs/ directory, these capabilities could be misused if the agent is misled by malicious input.
[PROMPT_INJECTION]: The skill ingests untrusted user data via the $ARGUMENTS variable and uses it to determine workflow behavior (dry-run vs. auto-fix) and as a parameter for shell execution. There are no boundary markers or explicit sanitization steps defined to prevent the agent from obeying instructions embedded within the arguments.
Ingestion points: $ARGUMENTS variable in SKILL.md.
Boundary markers: Absent.
Capability inventory: Local bash script execution (.specify/scripts/bash/create-cleanup.sh), directory moving, directory renaming, and file creation.
Sanitization: Absent; the skill relies on the agent's internal logic to parse arguments safely.

speckit-cleanup