The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill directs the agent to download resources from an external GitHub repository (github.com/dtran320/claud3) that does not belong to a verified or trusted organization.
[REMOTE_CODE_EXECUTION]: Instructions use npx skills add and /plugin install to download and install executable code and plugins from a remote source at runtime.
[COMMAND_EXECUTION]: The skill executes shell commands, including grep and test -f, to inspect the local filesystem and determine the installation status of the d3-viz skill.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes user-supplied data descriptions to determine chart types, which presents a surface for instructions embedded in data to influence the agent's behavior during code generation.
Ingestion points: User-provided data questions analyzed in SKILL.md.
Boundary markers: No explicit delimiters or boundary warnings are provided to separate data from instructions.
Capability inventory: Shell command execution, local file system reads, and dynamic generation of HTML/D3 code.
Sanitization: No sanitization or validation steps are documented for the incoming data questions.

claud3