linkfox-temu-promotion-global

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [CREDENTIALS_UNSAFE]: The onboarding.py utility contains a hardcoded Base64-encoded identifier (_LOGIN_FIXED_UID) that serves as a default authentication parameter when user-specific identifiers are missing.- [EXTERNAL_DOWNLOADS]: The skill requires several external Python libraries (requests, qrcode, pillow) for its onboarding and payment features, and provides instructions for their manual installation.- [COMMAND_EXECUTION]: The skill implements a comprehensive set of CLI scripts that manage local environment configuration, token storage, and business logic execution.- [DATA_EXFILTRATION]: All API request and response data, which may contain sensitive business information, is automatically recorded to local JSON files in the linkfox/ directory by the emit_result function in _temu_common.py.- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes and displays data retrieved from external Temu API endpoints.
  • Ingestion points: API responses received and parsed in _temu_common.py and secondary business scripts.
  • Boundary markers: The skill implements a response summarization strategy (_lf_summarize) to limit the volume of external data entering the model's context.
  • Capability inventory: File system writes (logging, token storage), network communication with vendor gateways, and command execution.
  • Sanitization: The skill validates JSON structures but does not explicitly sanitize the text content of API responses for malicious prompt patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 07:12 AM