The Agent Skills Directory

[SAFE]: The skill provides structured instructions and system prompts for an AI agent to act as a Technical Lead. It emphasizes architectural decision records (ADR), code review best practices (prioritizing correctness and security), and technical planning.
[INDIRECT_PROMPT_INJECTION]: As a code review tool, the skill is designed to ingest and analyze external code. While this presents a surface for indirect prompt injection, the skill includes explicit instructions for the agent to prioritize security and correctness during reviews, which serves as a basic defensive guideline. No evidence of exploitable logic was found.
[COMMAND_EXECUTION]: The Claude Code runtime configuration specifies permissions for bash, write, and edit. These are standard capabilities for that specific platform to allow the agent to perform its intended tasks, such as reviewing files and creating documentation. No malicious or hidden command sequences were detected in the instructions.

tech-leader-agent