The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests external design documents and developer-provided technical preferences, which are processed without explicit boundary markers or sanitization, creating a potential surface for indirect prompt injection.
1. Ingestion points: SKILL.md (Processing design documents and developer input).
1. Boundary markers: Absent.
1. Capability inventory: Writing documentation, creating project files, environment initialization, and dependency installation.
1. Sanitization: Absent.
[COMMAND_EXECUTION]: The skill automates project initialization and dependency installation based on the confirmed technical scheme, which is a core functional requirement of its stated purpose.
[SAFE]: The skill follows secure secret management practices by instructing the agent to create .env.example files and explicitly warning that sensitive values must not be submitted to the repository.

dev-planning