The Agent Skills Directory

[COMMAND_EXECUTION]: Executes a local shell script .specify/scripts/bash/check-prerequisites.sh to initialize the environment and extract project metadata.\n- [DYNAMIC_EXECUTION]: Implements a hook system that resolves command names from the .specify/extensions.yml configuration file and triggers them using the EXECUTE_COMMAND directive. This allows project-level configuration to dynamically invoke agent commands, which can lead to unintended tool execution if the configuration file is modified by an attacker.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes content from multiple external files to drive its logic and output.\n
Ingestion points: Reads .specify/extensions.yml for extension hooks, and spec.md, plan.md, and tasks.md for feature context loading.\n
Boundary markers: The instructions do not define boundary markers or explicit delimiters to isolate the untrusted project file content from the agent's instructions.\n
Capability inventory: The skill can execute local scripts, invoke dynamic agent commands via the hook system, and write new markdown files to the checklists/ directory.\n
Sanitization: There is no evidence of sanitization or validation for the command strings extracted from the configuration or the text parsed from requirement documents.

speckit-checklist