doubao-ask

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads an environment setup script from the vendor's official GitHub repository.
  • Evidence: scripts/setup.sh fetches content from https://raw.githubusercontent.com/FinpeakInc/frevana-cli-releases/refs/heads/main/skills/frevana/scripts/setup.sh.
  • [COMMAND_EXECUTION]: The skill executes downloaded scripts and local binaries to interact with the Frevana daemon.
  • Evidence: scripts/setup.sh uses exec bash to run the downloaded content. scripts/ask_doubao.sh executes the frevana binary installed in the user's home directory.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes input from local files without sanitization or boundary markers.
  • Ingestion points: ask_doubao.sh reads prompts from a user-specified file via the --prompt-file argument.
  • Boundary markers: Absent. No delimiters are used to separate user data from instructions.
  • Capability inventory: The skill can perform network operations via the local Frevana daemon and execute shell commands.
  • Sanitization: Absent. Content from the prompt file is passed directly to the execution pipeline.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 03:39 PM
Security Audit — agent-trust-hub — doubao-ask