The Agent Skills Directory

[COMMAND_EXECUTION]: The skill runs a local PowerShell script located at .specify/scripts/powershell/check-prerequisites.ps1. This is used to initialize the environment and extract directory metadata, which is consistent with the skill's documented prerequisite for the 'spec-kit' framework.- [PROMPT_INJECTION]: The skill is subject to potential indirect prompt injection as it ingests untrusted data from local files.
Ingestion points: Reads documentation from spec.md, plan.md, and tasks.md within the FEATURE_DIR.
Boundary markers: The instructions lack explicit XML or unique delimiters for the ingested content, though they provide strict constraints on the agent's output logic.
Capability inventory: Includes the ability to execute shell scripts and write new markdown files to the local file system.
Sanitization: No specific sanitization or filtering is performed on the ingested documentation content. Despite these surface risks, the skill includes robust instructions and negative examples (anti-patterns) to ensure the agent remains focused on requirement quality validation rather than implementation verification.

speckit-checklist