The Agent Skills Directory

[SAFE]: The skill is entirely instructional and focused on prompt 'tree-shaking'—the removal of redundant or unnecessary instructions to improve efficiency.
[NO_CODE]: No executable scripts, binaries, or configuration files that trigger automation were found in the skill.
[DATA_EXPOSURE_AND_EXFILTRATION]: No network requests or access to sensitive local files (e.g., credentials, SSH keys) are present in the instructions.
[INDIRECT_PROMPT_INJECTION]: The skill processes workspace prompt files which represents a vulnerability surface, however, it lacks active capabilities like code execution or network operations to be exploited.
Ingestion points: Workspace system prompts, skill files, and instruction docs.
Boundary markers: None specified.
Capability inventory: Text analysis and reporting recommendations.
Sanitization: None specified.

melech-prompt-shake