The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external websites and URL inventories to generate reports and mapping documents, which creates a potential surface for indirect prompt injection.
Ingestion points: Content retrieved from external domains via scripts/dfs_client.py crawl and scripts/site_crawler.py, as well as user-provided URL mapping CSV files.
Boundary markers: The skill does not define specific delimiters or instructions to ignore potentially malicious content embedded in the crawled data.
Capability inventory: The skill invokes multiple local Python scripts (dfs_client.py, drift_store.py, site_crawler.py, google_client.py, report_publish.py) and performs file writing operations to the local filesystem.
Sanitization: No explicit content sanitization or validation logic is mentioned for the data fetched from remote URLs.
[COMMAND_EXECUTION]: The workflow relies on executing local Python scripts located in the scripts/ directory. These scripts are invoked via the command line with user-supplied arguments such as domains and keywords, which is a standard operational pattern but constitutes the skill's primary capability set.

workflow-migration