The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The script champion_tracker.py dynamically loads an external Python module (enrich_leads.py) from a relative path in another skill's directory (lead-qualification/scripts/) using importlib.util. This is used for cross-skill code reuse but involves runtime execution of code from computed paths.
[EXTERNAL_DOWNLOADS]: The skill uses the requests library to communicate with app.gooseworks.ai (the author's official domain) to proxy enrichment requests to Apify. This is a documented core feature for processing LinkedIn profiles.
[COMMAND_EXECUTION]: The Python script performs file system operations, including searching for and reading .env files in parent directories and writing results to the snapshots/ and output/ folders.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface through the ingestion of external data.
Ingestion points: Processes data from champions.csv and external profile enrichment data.
Boundary markers: None present; the script treats CSV content as trusted data for scoring and reporting.
Capability inventory: Network access via requests and file-writing capabilities across multiple directories.
Sanitization: No sanitization is performed on input data beyond normalization of URLs and company names.

champion-tracker