The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external data by fetching source code and integration examples from the expo/examples repository on GitHub. This creates a potential surface for indirect prompt injection where instructions could be hidden in repository content to influence agent behavior.
Ingestion points: Content is retrieved using gh api, curl, and git clone from github.com/expo/examples (SKILL.md).
Boundary markers: The instructions do not specify explicit delimiters or warnings to ignore instructions embedded within the fetched code files.
Capability inventory: The skill allows the execution of gh, git, npx, bun, and curl via the shell (SKILL.md).
Sanitization: There is no explicit sanitization of the fetched external content before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill references and downloads content from external sources, primarily targeting the official Expo GitHub organization and well-known registries. These interactions are consistent with the skill's purpose.
The gh api, git clone, and npx degit commands target the github.com/expo/examples repository.
The npx create-expo and bun create expo commands download project scaffolding tools.
[COMMAND_EXECUTION]: The skill facilitates the execution of shell commands to manage project files and dependencies. This includes using gh for GitHub API interactions, git for repository cloning, and npx for running project creation and feedback tools.

expo-examples