Skills
Navigate
skills/sqrdao/sqrskills/telegram-group-summary/Gen Agent Trust Hub

telegram-group-summary

Pass

Audited by Gen Agent Trust Hub on Sep 27, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-generated content from Telegram messages.
  • Ingestion points: The fetch_messages.py script reads data from local JSON/JSONL files in OPENCLAW_STATE_DIR and from the Telegram Bot API getUpdates endpoint.
  • Boundary markers: The instructions in SKILL.md do not specify the use of delimiters (e.g., XML tags or triple backticks) or "ignore embedded instructions" warnings when passing the raw messages to the agent for summarization.
  • Capability inventory: The skill is allowed to use Bash(python3 *), giving it the ability to execute code based on context.
  • Sanitization: The _normalize_message function in scripts/fetch_messages.py extracts the text and caption fields without any filtering or sanitization for malicious prompt instructions.
  • [COMMAND_EXECUTION]: The skill executes Python scripts via the Bash tool to perform its operations.
  • Evidence: SKILL.md lists Bash(python3 *) in allowed-tools and provides instructions to run scripts/fetch_messages.py and list_groups.py via shell commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 27, 2026, 09:05 AM