The Agent Skills Directory

[PROMPT_INJECTION]: No malicious instruction overrides or safety bypass attempts were detected. The instructions are functional and restricted to linguistic reformatting.
[DATA_EXPOSURE]: The skill references internal repository files like CONTEXT.md and CONTEXT-MAP.md for terminology guidance. It does not attempt to access sensitive system files, environment variables, or credentials.
[COMMAND_EXECUTION]: The skill does not contain shell commands or script execution. The frontmatter explicitly sets disable-model-invocation: true, preventing the agent from calling external tools while this skill is active.
[EXTERNAL_DOWNLOADS]: No remote scripts, packages, or external resources are referenced or downloaded.

wait-what