The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a local shell script located at .specify/scripts/bash/check-prerequisites.sh. This script is used to gather environment context (FEATURE_DIR and AVAILABLE_DOCS) in JSON format. The instructions include specific guidance on escaping single quotes to prevent shell argument injection.
[DATA_EXPOSURE]: The workflow involves reading local project files such as spec.md, plan.md, and tasks.md. Access is restricted to the directory defined by the FEATURE_DIR variable discovered during setup, which is consistent with the skill's purpose of requirement analysis.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect injection surface because it processes untrusted content from external markdown files (spec.md, plan.md, etc.) to generate questions and checklist items.
Ingestion points: Files spec.md, plan.md, and tasks.md located within the FEATURE_DIR (SKILL.md, Workflow Step 4).
Boundary markers: None explicitly defined for document processing; however, the skill employs a "Load only necessary portions" strategy to limit context ingestion.
Capability inventory: Shell script execution (check-prerequisites.sh), directory creation (mkdir), and markdown file writes (checklists/[domain].md).
Sanitization: The skill provides specific shell escaping logic for user arguments but does not explicitly document sanitization for content read from project files.

speckit-checklist