The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes external data from Temu's APIs (via the LinkFox gateway), which could potentially contain malicious instructions intended to influence the agent.
Ingestion points: Responses from API calls to tool-gateway.linkfox.com are loaded and printed to the session's stdout in scripts/_temu_common.py and scripts/_temu_global_common.py.
Boundary markers: The skill does not implement specific delimiters or warnings to ignore instructions embedded in the API data.
Capability inventory: The skill has capabilities for network access (API requests), local file system writes (session logs), and reading local configuration files.
Sanitization: API response data is printed directly or summarized without specific sanitization of potentially executable text.
[DATA_EXFILTRATION]: The skill accesses a sensitive local file path at ~/.linkfox/temu-access-tokens.json to store and retrieve Temu access tokens.
Evidence: scripts/_temu_token_store.py defines DEFAULT_STORE_PATH targeting the user's home directory for token persistence.
Note: The network exfiltration of these tokens to *.linkfox.com is considered normal vendor behavior as the domain belongs to the skill's author.
[COMMAND_EXECUTION]: The skill provides a suite of Python scripts for managing Temu orders, including authentication guides and file downloaders, which perform network operations and file system manipulations.
Evidence: scripts/_temu_common.py contains logic to search for and create directories in getcwd(), ~, and tempdir for session data storage.

linkfox-temu-cancel-order-global