skills/yerza06/tguser-cli/tguser/Gen Agent Trust Hub

tguser

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
  • [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill facilitates the installation of the tguser CLI tool from a remote GitHub repository.
  • Evidence: uv tool install git+https://github.com/yerza06/tguser-cli.git in SKILL.md.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill provides capabilities to send text, media, and files from the local filesystem to Telegram, which is an external network destination.
  • Evidence: Usage of tguser send, tguser sendfile, and tguser sendphoto commands to transmit data to external recipients.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data (user-provided messages and file contents) and possesses the capability to exfiltrate this data via network operations using shell commands.
  • Ingestion points: User-defined message text and file paths processed by the tguser commands in SKILL.md.
  • Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within the content being sent.
  • Capability inventory: Invokes the tguser CLI to perform network-based transmission of data via shell execution.
  • Sanitization: The instructions do not specify any sanitization or validation of the content before it is passed to the Telegram API.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 07:51 PM