The Agent Skills Directory

[COMMAND_EXECUTION]: The skill facilitates the execution of local development commands such as npm test, npm run lint, tsc, and git commands as part of its primary software development workflow.
[PROMPT_INJECTION]: The skill uses instruction-heavy templates to guide agent behavior. It also contains an indirect prompt injection surface where it ingests external data (ingestion points: Requirements Checklist, Success Criteria in SKILL.md). It utilizes markdown headers and blockquotes as boundary markers. Capability inventory includes Read, Write, Edit, and Bash tools. No explicit sanitization of input data is defined.
[DATA_EXFILTRATION]: The skill references reading configuration files like .env.example and tsconfig.json. These actions are limited to local file access for context gathering and do not involve unauthorized network transmissions.

planning-patterns