pp-tesla

Warn

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Skill requires installation of tesla_auth from a third-party source (adriankumpf) via Cargo or Homebrew to handle OAuth login.
  • [EXTERNAL_DOWNLOADS]: Fetches and builds tesla-control from Tesla's official GitHub repository to support signed commands for 2021+ vehicles.
  • [DATA_EXFILTRATION]: Supports routing command output to arbitrary external URLs via the --deliver webhook:<url> flag.
  • [DATA_EXFILTRATION]: Includes a feedback mechanism that can transmit data to a remote endpoint if TESLA_FEEDBACK_ENDPOINT is configured and auto-send is enabled.
  • [COMMAND_EXECUTION]: Requests Read Bash capability to execute the Tesla CLI, manage authentication state, and perform vehicle commands.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: Fetches real-time vehicle state, release notes, and service data from the Tesla API.
  • Boundary markers: No explicit boundary markers or instructions to ignore embedded content are provided for interpolated data.
  • Capability inventory: Can execute powerful commands via Read Bash, including unlocking vehicles, remote starting, and managing security keys.
  • Sanitization: No evidence of output sanitization or validation of the remote API content before presenting it to the agent context.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 14, 2026, 11:07 PM