The Agent Skills Directory

[NO_CODE]: The skill package does not contain any executable scripts, binaries, or code files. It consists exclusively of natural language instructions, configuration, and documentation.
[SAFE]: The analysis found no evidence of malicious behavior. The instructions are transparently designed for prompt refinement, and the configuration fields ('disable-model-invocation', 'opencode/autoinvoke') appropriately limit the skill's execution environment. No prompt injections, data exfiltration attempts, or obfuscation techniques were identified.

sharpen