The Agent Skills Directory

[SAFE]: The skill provides legitimate technical documentation for GLSL shaders in the Defold engine. No malicious patterns were detected across any threat categories.
[PROMPT_INJECTION]: No evidence of prompt injection attempts, safety filter bypasses, or instructions to ignore previous system prompts were found.
[DATA_EXFILTRATION]: No sensitive file paths, hardcoded credentials, or network operations targeting external domains were identified.
[OBFUSCATION]: The content was analyzed for Base64, zero-width characters, homoglyphs, and hidden URLs; no obfuscated content was found.
[REMOTE_CODE_EXECUTION]: The skill does not perform any remote package installations or script executions. It focuses entirely on local shader file editing.
[COMMAND_EXECUTION]: No dangerous shell commands or privilege escalation attempts (such as sudo or chmod) were detected.
[INDIRECT_PROMPT_INJECTION]: While the skill assists in code generation, it does not ingest untrusted external data in a way that poses an injection risk to the agent environment.
[DYNAMIC_CONTEXT_INJECTION]: No !command execution syntax was found in the SKILL.md file.

defold-shaders-editing