The Agent Skills Directory

[SAFE]: The skill contains no malicious code, unauthorized network access patterns, or attempts to access sensitive system data. It serves exclusively as a style guide and auditing tool for the agent.- [NO_CODE]: All 12 files in the skill are markdown documents. No Python scripts, Node.js packages, or binary executables are included, eliminating risks associated with remote code execution or package dependencies.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze untrusted text strings (UI copy). While this creates an ingestion point for potentially malicious instructions, the skill has no functional capabilities (such as writing to files or making network requests) that could be used to execute an attack. 1. Ingestion points: Analyzes 'every visible string in the deliverable' as specified in the Steps section of SKILL.md. 2. Boundary markers: No explicit delimiters or boundary warnings are present in the instructions. 3. Capability inventory: None; the skill does not include any scripts or executable tools. 4. Sanitization: No sanitization or filtering of input copy strings is performed.

ux-copy