The Agent Skills Directory

[SAFE]: The skill operates as a high-level orchestrator using sequential and parallel task logic. It does not perform sensitive operations itself but coordinates specialized sub-skills. No hardcoded credentials, malicious commands, or exfiltration patterns were detected.
[PROMPT_INJECTION]: The skill manages an attack surface for indirect prompt injection as it ingests untrusted requirements and codebases, but it includes appropriate mitigations. Ingestion points: Processes user requirements and existing codebase patterns in Phase 1 (Analysis) and Phase 2 (Design) within SKILL.md. Boundary markers: Data isolation is handled through state-driven phase transitions and specific output objects for each skill invocation. Capability inventory: Coordinates skills for code generation and test execution, which are high-capability tools. Sanitization: Implements a dedicated Review Phase (Phase 5) that utilizes a security-reviewer skill to perform security analysis and enforce a quality gate requiring no security vulnerabilities.

coding-workflow