The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Fetches source code and tool configuration from the project's repository at https://github.com/nexu-io/open-design.
[COMMAND_EXECUTION]: Utilizes Bash tools to clone repositories, manage Node.js environment via corepack, and control the tools-dev daemon process (start, stop, status).
[REMOTE_CODE_EXECUTION]: Downloads and executes code during the installation process through pnpm install and the addition of external skills via npx skills add from the author's repository (https://github.com/akillness/jeo-skills).
[PROMPT_INJECTION]: Identifies an indirect prompt injection surface where user-provided design descriptions are processed by local coding agents to generate artifacts.
Ingestion points: User prompts captured in Step 1 for UI prototypes, decks, and documents as described in SKILL.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the prompt interpolation logic.
Capability inventory: The execution environment has access to powerful tools including Bash, Write, Edit, Glob, and Grep as defined in the skill's frontmatter.
Sanitization: There is no evidence of sanitization or validation of external prompt content before it is passed to the underlying agents.

open-design