The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to process and manipulate content dictionary files (e.g., *.content.ts, *.content.json). These files, which may contain user-provided data, represent an ingestion surface where malicious instructions could theoretically be embedded to influence the agent's behavior during content reviews or migrations.
Ingestion points: The skill reads from files defined in the content.contentDir (specified in references/content-dictionaries.md and references/setup-tanstack-start.md).
Boundary markers: No specific boundary markers or instruction isolation techniques are mentioned for handling these content files.
Capability inventory: The skill allows for command execution via bunx, file modification, and network synchronization through intlayer push/pull commands.
Sanitization: There is no evidence of sanitization or filtering of the content within the dictionary files before they are processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill provides instructions for installing dependencies from the npm registry, such as intlayer, react-intlayer, and vite-intlayer. These are standard packages required for the tool's core functionality.
[COMMAND_EXECUTION]: The skill references multiple CLI commands for project initialization, building, and synchronization (e.g., bunx intlayer build, bunx intlayer push). These commands are standard for the Intlayer development workflow and are executed in the context of the developer's project environment.

intlayer