The Agent Skills Directory

[SAFE]: The skill operates entirely within the local design environment. Analysis confirms there are no network operations, external downloads, or attempts to access sensitive system files or credentials.
[COMMAND_EXECUTION]: The skill uses execute_code to run provided JavaScript snippets for design tree traversal and layer property updates. These scripts are statically defined within the skill and do not incorporate untrusted external input into executable logic.
[PROMPT_INJECTION]: Although the skill processes untrusted data from design files (such as layer names and text content), the risk of indirect prompt injection is mitigated by a multi-phase workflow. The agent is instructed to perform read-only discovery and inference, presenting a full plan for human approval before executing any modifications.

penpot-rename-layers