The Agent Skills Directory

[DATA_EXFILTRATION]: The skill handles Personally Identifiable Information (PII), including customer names, phone numbers, email addresses, and shipping locations (e.g., in bg.order.decryptshippinginfo.get). This data is retrieved from Temu and transmitted through the vendor-controlled gateway at tool-gateway.linkfox.com. Additionally, scripts/onboarding.py transmits user phone numbers to api.linkfox.com during the account registration process.
[CREDENTIALS_UNSAFE]: The skill manages Temu access tokens by storing them in a local JSON file at ~/.linkfox/temu-access-tokens.json via scripts/_temu_token_store.py. While intended for functionality, these files contain sensitive credentials that grant access to the user's Temu store data.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external orders (e.g., customization text, recipient names) and writes this data to local JSON files in the project directory. If the AI agent subsequently reads these files, there is a risk of indirect prompt injection if an order contains malicious instructions designed to influence the agent's behavior.
Ingestion points: Data is fetched from Temu APIs (e.g., bg.order.detail.v2.get, bg.order.customization.get) and stored in <cwd>/linkfox/.
Boundary markers: None identified in the script outputs or documentation.
Capability inventory: The skill can perform network requests to the vendor gateway and write files to the local filesystem.
Sanitization: No explicit sanitization or filtering of order content was identified in the provided scripts.
[COMMAND_EXECUTION]: The skill provides numerous Python scripts in the scripts/ directory meant to be executed by the agent to interact with the API. This requires the agent to have permission to execute subprocesses in the local environment.

linkfox-temu-order-us