The Agent Skills Directory

[SAFE]: The skill is purely instructional and provides a manual workflow for developers and AI agents to refactor code duplication. It does not contain any executable scripts, unexpected network calls, or hidden logic.
[NO_CODE]: The skill contains markdown documentation and workflow instructions but does not ship with any Python or JavaScript executable files or tool definitions.
[EXTERNAL_DOWNLOADS]: The instructions describe using npx jscpd to identify code clones. This refers to an established open-source tool maintained by the skill's author for code analysis purposes.
[COMMAND_EXECUTION]: The workflow instructs the use of npx jscpd with various CLI flags to analyze the local codebase for duplication hotspots.
[INDIRECT_PROMPT_INJECTION]: The workflow requires the agent to read and process source code from the local codebase to identify and refactor clones.
Ingestion points: The agent is instructed to read file contents and line ranges identified as clones into its context (Workflow step 3).
Boundary markers: The skill does not specify any specific delimiters or instructions to ignore potential commands embedded within the source code files being read.
Capability inventory: The workflow guides the agent to perform file reads, file writes (to apply refactorings), and shell command execution via the terminal.
Sanitization: The skill does not define methods for sanitizing or filtering the ingested source code before it is processed by the agent.

dry-refactoring