skills/mtgo-labs/mtgo/mtgo-mtproto-go/Gen Agent Trust Hub

mtgo-mtproto-go

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The library invokes the ffmpeg system utility in telegram/ffmpeg_pipe.go to facilitate multimedia broadcasting. The command is constructed using a slice of arguments, which is a safe practice that prevents shell-based command injection.
  • [PROMPT_INJECTION]: The skill represents an attack surface for indirect prompt injection because it processes untrusted messages from the Telegram API. Ingestion points: Update handling logic in telegram/client.go. Boundary markers: None; the library provides raw message text and entities to the application layer. Capability inventory: Significant capabilities including chat administration, message deletion, and media streaming via ffmpeg. Sanitization: Incoming data is structured into TL types, but natural language instructions from users are not filtered by the library.
  • [SAFE]: The project structure, license, and external dependencies are standard for a Go-based API client. The go.mod file contains a future version number (1.26.2), which is likely a non-malicious configuration placeholder or error.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 05:50 PM
Security Audit — agent-trust-hub — mtgo-mtproto-go