The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the script .specify/scripts/bash/create-hotfix.sh --json "$ARGUMENTS" using unsanitized user input from the $ARGUMENTS variable. This allows an attacker to execute arbitrary shell commands by including shell metacharacters (e.g., ;, &&, |, or backticks) in their input.
[PROMPT_INJECTION]: The skill contains instructions that explicitly mandate bypassing standard safety and quality protocols, referring to them as a "deviation from the constitution." This is a pattern used to override the agent's core instructions and safety filters.
[PROMPT_INJECTION]: Indirect prompt injection surface identified. \n
Ingestion points: $ARGUMENTS variable in SKILL.md. \n
Boundary markers: Absent; no delimiters or instructions to ignore embedded commands. \n
Capability inventory: Shell command execution and file system writes. \n
Sanitization: None; raw user input is interpolated directly into a shell command.

speckit-hotfix