The Agent Skills Directory

[SAFE]: The skill defines a structured protocol for critical thinking and evidence-based decision-making, instructing agents to treat external feedback as hypotheses to be verified through independent investigation.
[EXTERNAL_DOWNLOADS]: The skill incorporates mandatory web searches as a verification mechanism to ground technical assessments in current documentation and community best practices.
[COMMAND_EXECUTION]: The instructions direct the agent to use standard tools for codebase inspection, such as file read and search operations, to verify claims against the actual project implementation.
[SAFE]: The included reference file for AI false positive patterns enhances security by training the agent to recognize common patterns of over-engineering or symptomatic fixes that do not address root causes.
[SAFE]: No malicious patterns, such as prompt injection, obfuscation, or unauthorized data exfiltration, were identified across the analyzed files.

assess-findings