The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill downloads the @tanstack/intent package from the official NPM registry. This resource belongs to a well-known and trusted organization in the web development ecosystem.
[REMOTE_CODE_EXECUTION]: The skill uses bunx to execute remote code from the @tanstack/intent package at runtime, which is the intended behavior for loading documentation and guidance.
[COMMAND_EXECUTION]: Instructs the agent to execute shell commands for listing, loading, and installing package guidance and repository hooks.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the local project environment which could influence subsequent command parameters.
Ingestion points: The agent reads the local package.json file and the JSON output of the intent list command as described in SKILL.md.
Boundary markers: No specific delimiters or boundary instructions are used when processing these data sources.
Capability inventory: The skill performs shell command execution using bunx in both SKILL.md and SETUP.md.
Sanitization: No explicit sanitization logic for ingested package names or IDs is provided within the skill instructions.

tanstack-intent