The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes multiple Python scripts that allow the agent to perform administrative tasks on Temu Ads, such as creating campaigns, querying logs, and generating reports.
[CREDENTIALS_UNSAFE]: The _temu_token_store.py script manages the local storage of Temu access tokens in the ~/.linkfox/temu-access-tokens.json file. This is an intended mechanism for preserving session authentication within the user's local environment.
[DATA_EXFILTRATION]: Network operations in scripts/_temu_common.py and scripts/onboarding.py target vendor-controlled subdomains of linkfox.com and official temu.com endpoints. These communications are required for core functions such as gateway proxying and account onboarding.
[PROMPT_INJECTION]: The skill processes external data from the Temu API, which presents a surface for indirect prompt injection.
Ingestion points: External data is ingested via the call_temu_api function in scripts/_temu_common.py.
Boundary markers: There are no explicit delimiters or safety instructions used to isolate API response data from the agent's instructions.
Capability inventory: The agent has the ability to modify ad campaigns and budgets via scripts like eu_ads_searchrec_ad_modify.py.
Sanitization: Response data is printed to the console using emit_result without specific filtering for malicious instruction patterns.

linkfox-temu-ads-eu