The Agent Skills Directory

COMMAND_EXECUTION (HIGH): The skill's workflow automatically executes a local bash script .specify/scripts/bash/check-prerequisites.sh --json during the setup phase. Executing scripts from the skill directory can lead to arbitrary code execution on the user's machine if the script content is malicious or modified.
PROMPT_INJECTION (HIGH): The skill is vulnerable to Indirect Prompt Injection (Category 8) due to its high-privilege capabilities combined with external data ingestion.
Ingestion points: The skill reads external files including spec.md, plan.md, and tasks.md to load context.
Boundary markers: Absent. The instructions do not define clear delimiters or instruct the agent to ignore instructions embedded within the loaded context files.
Capability inventory: The skill has the capability to execute bash scripts (via check-prerequisites.sh) and write new files to the filesystem (FEATURE_DIR/checklists/[domain].md).
Sanitization: Absent. There is no evidence of sanitization, filtering, or validation of the content read from the markdown files before it is used to influence the agent's behavior or file-writing operations.

spec-checklist