The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch canonical documentation and inspect script implementations from an external GitHub repository (github.com/nlweb-ai/NLWeb). This is used to verify CLI flags and mapping logic before execution.
[COMMAND_EXECUTION]: The skill uses the Bash tool to run the local Python module data_loading.db_load. These commands handle data ingestion, site partitioning, and database cleanup operations.
[PROMPT_INJECTION]: The skill establishes a pipeline for ingesting untrusted third-party data (RSS/Atom feeds, Schema.org JSON-LD, and CSVs) which presents an indirect injection surface.
Ingestion points: External URLs provided via command line or URL list files (--url-list).
Boundary markers: None mentioned in the skill instructions; the agent is not instructed to use specific delimiters when processing this data.
Capability inventory: Bash (command execution), WebFetch (data retrieval), Read/Write (configuration and database access).
Sanitization: No explicit sanitization or validation steps are outlined in the skill instructions, relying on the internal logic of the db_load.py script.

nlweb-data-loading