tg-rich-streaming
Pass
Audited by Gen Agent Trust Hub on Aug 8, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill communicates exclusively with the official Telegram Bot API at
api.telegram.org. This is a well-known service and the expected behavior for a Telegram integration. - [DATA_EXFILTRATION]: The skill requires a
TELEGRAM_BOT_TOKEN. The provided demo script correctly instructs users to provide this via an environment variable (TELEGRAM_BOT_TOKEN), which is a standard and safe practice for secret management. - [COMMAND_EXECUTION]: The
scripts/stream_demo.pyis a standalone simulation script. It uses standard Python libraries (urllib,json,argparse) to demonstrate the API workflow. It does not execute arbitrary shell commands or access sensitive local files. - [PROMPT_INJECTION]: The skill documentation defines a UX pattern for streaming. While it handles AI-generated text, it does not provide instructions that attempt to bypass safety filters or override agent behavior.
- [INDIRECT_PROMPT_INJECTION]: The skill acts as an output interface for LLM data. While it processes untrusted content (LLM chunks), it does so within the context of a Telegram message display. The risk of indirect injection is low as the data is passed to a third-party API (Telegram) rather than being executed locally or used to influence subsequent agent reasoning steps.
Audit Metadata