The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute standard development commands (such as git, test runners, and build tools) to provide evidence for status claims. These commands are part of the intended purpose of verifying code integrity and are accompanied by strict protocols for analyzing output logs.
[SAFE]: The skill establishes 'Verification Gates' and feedback reception protocols that serve as security best practices. By requiring technical skepticism and evidence-based verification of external input, the skill inherently mitigates risks associated with indirect prompt injection from external reviewers.
[PROMPT_INJECTION]: The skill contains directives to avoid 'performative agreement' and social filler words (e.g., 'Thanks', 'You are right'). These are operational constraints designed to ensure technical rigor and do not attempt to bypass core safety guardrails or safety-aligned behaviors.

code-review