The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute several project-specific commands using bun to verify changes, including bun test:all, bun generate:all, and bun --filter @seed-design/docs build:lynx-examples:development. These are standard development operations for the project but involve direct shell interaction.
[INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where the agent must read AGENTS.md files from multiple local directories (docs/, docs/examples/lynx/, etc.) and fetch compatibility JSON data from external GitHub repositories (github.com/lynx-family/lynx-website). This creates a surface where malicious content in these files or remote data could override the agent's instructions.
Ingestion points: Reads local AGENTS.md files, React documentation files (.mdx), and remote compatibility data from GitHub.
Boundary markers: None explicitly defined to separate untrusted data from the agent's core instructions.
Capability inventory: The agent has the ability to write to the file system (creating/modifying documentation and examples) and execute shell commands via bun.
Sanitization: There is no mention of sanitizing or validating the content of the ingested files before processing or acting upon them.

write-lynx-component-docs