The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local shell script scripts/check-prerequisites.sh to scan a user-provided directory for specific feature artifacts. The script uses jq to safely handle the directory path in its JSON output.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to read, analyze, and "internalize" content from various markdown files (e.g., spec.md, ux.md, plan.md) to build a context for the agent.
Ingestion points: Multiple project files including spec.md, ux.md, plan.md, tasks.md, data-model.md, research.md, and setup.md located in the directory passed to the skill.
Boundary markers: Absent. The instructions do not provide delimiters or specific commands to treat the content of these files as data rather than instructions.
Capability inventory: The skill can execute local shell scripts and read files from the local filesystem.
Sanitization: While the directory path is sanitized for JSON output in the helper script, the actual content of the documentation files is ingested directly into the agent's context without validation or sanitization, potentially allowing embedded instructions in those files to influence agent behavior.

feature-analyzer