The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external, untrusted data which could contain malicious instructions designed to override agent behavior or influence its outputs.
Ingestion points: Documentation retrieved via the tanstack search-docs CLI tool, content from community skills (e.g., DeckardGer/tanstack-agent-skills), and GitHub-hosted resources such as issues and releases for the TanStack ecosystem (as referenced in references/current-authority.md).
Boundary markers: Absent. The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when processing this external content.
Capability inventory: The skill utilizes command execution through the tanstack CLI, the opensrc utility, and local maintenance scripts such as tools/skill/quick_validate.py (referenced in SKILL.md and references/current-authority.md).
Sanitization: Absent. There is no guidance provided for sanitizing, validating, or filtering the content retrieved from external sources before it is processed by the agent.
[COMMAND_EXECUTION]: The skill relies on the execution of several CLI tools and local scripts as part of its documented workflow for maintenance and information retrieval.
Evidence: The instructions suggest using the tanstack CLI for documentation lookups, the opensrc tool for inspecting package source paths, and python3 for running local validation and checking scripts (found in SKILL.md and references/current-authority.md).

tanstack-router