telegram-formatting
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a formatting guide for an AI agent. Analysis of the instructions and reference materials found no evidence of prompt injection, credential theft, or unauthorized network operations.
- [INDIRECT_PROMPT_INJECTION]: The skill outlines the delivery of content to external users via Telegram, which constitutes an ingestion surface for potentially untrusted data. However, the documentation focuses on output formatting and explicitly states that the platform gateway handles character escaping and conversion.
- Ingestion points: Text content generated or structured by the agent for delivery via the Telegram gateway (SKILL.md).
- Boundary markers: The skill does not prescribe specific boundary delimiters for user-provided data within the formatted output (SKILL.md).
- Capability inventory: Rich rendering triggers including tables, checklists, collapsible details blocks, math formulas, media embedding, and file delivery via local paths (SKILL.md, adapter-internals.md).
- Sanitization: Documentation specifies that the platform gateway automatically handles conversion and escaping for the legacy MarkdownV2 path (SKILL.md).
Audit Metadata