The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to fetch live documentation and configuration files from a remote repository and use that information to perform setup tasks and file edits. This ingestion of external data creates a surface for indirect prompt injection.
Ingestion points: The agent is directed to fetch README, nlweb-hello-world.md, nlweb-cli.md, sites.xml, and tools.xml from github.com/nlweb-ai/NLWeb.
Boundary markers: Absent. The instructions do not specify delimiters or provide explicit warnings to ignore potential instructions embedded in the fetched content.
Capability inventory: The skill utilizes Write, Edit, and Bash tools to modify the local filesystem and execute setup commands based on the fetched data.
Sanitization: Absent. There is no evidence of filtering or validation for the content retrieved from the external documentation before it influences agent actions.
[EXTERNAL_DOWNLOADS]: The skill fetches documentation, configuration templates, and project release notes from the nlweb-ai/NLWeb repository on GitHub to guide the setup process.
[COMMAND_EXECUTION]: The skill invokes the nlweb CLI tool for environment initialization (nlweb init-python, nlweb init, nlweb check) and uses standard shell commands to clone the repository and manage Python virtual environments.

nlweb-setup