The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes structured JSON responses from external APIs (Amazon and LinkFox) at runtime.
Ingestion points: Data is ingested through API responses in scripts/authorize_url.py, scripts/authorized_stores.py, scripts/refresh_token.py, scripts/store_tokens.py, and scripts/onboarding.py.
Boundary markers: The skill uses specific JSON parsing and result extraction logic (such as the _lf_summarize function in scripts/_lf_output.py) to delimit and summarize responses.
Capability inventory: The skill performs network requests via urllib and requests, and writes session metadata and data files to the linkfox/ directory in the project root.
Sanitization: Responses are parsed as JSON; additionally, scripts/_token_status_output.py explicitly strips sensitive accessToken and refreshToken fields from result objects before they are logged or displayed.
[EXTERNAL_DOWNLOADS]: The scripts/onboarding.py script references standard Python libraries qrcode, pillow, and requests. It provides instructions for the user to install them via pip if they are not already available in the environment.
[COMMAND_EXECUTION]: The documentation in references/onboarding.md provides manual shell commands (export, setx) for the user to configure their API key in the environment. These commands are for user guidance and are not executed automatically by the agent scripts.

linkfox-amazon-store-auth