The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze local component code, hooks, and queries to map application states and generate a development workbench.
Ingestion points: Project source code (components and hooks) is ingested by the agent during the scoping and mapping sequence described in SKILL.md.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions when processing the ingested source code.
Capability inventory: The agent has the capability to write new files (React/framework pages, fixtures) and modify project routing configurations across multiple frameworks as detailed in SKILL.md and references/switcher.md.
Sanitization: No explicit sanitization or validation of the analyzed code is requested before it is used to generate the workbench files.
[DYNAMIC_EXECUTION]: The skill involves the generation of new functional scripts and web pages derived from project templates and existing code.
Evidence: SKILL.md (Step 3) and references/switcher.md instruct the agent to create new pages (/lab/states/<component>) that import and execute components in a local environment.
[COMMAND_EXECUTION]: The skill requires the agent to perform file system maintenance and run development-related commands.
Evidence: SKILL.md (Step 6) directs the agent to delete specific folders, remove lines from route files, and execute type-check or build commands to verify the cleanup process.

state-machine