apigram-mcp
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes an interface for the agent to interact with Telegram, which exposes it to untrusted data from external sources (Telegram users and channels).
- Ingestion points: Tools such as
list_dialogs,get_chat, andget_historydescribed inSKILL.mdingest chat metadata and message history from external sources into the agent's context. - Boundary markers: The documentation does not specify the use of delimiters or specific instructions to help the agent distinguish between system instructions and content within the retrieved messages.
- Capability inventory: The skill provides multiple tools to modify state or interact with the environment, such as
send_message,edit_message,delete_messages,react, andforward_messages, which could be targeted by instructions embedded in incoming messages. - Sanitization: No sanitization or validation mechanisms are described for the content retrieved from Telegram before it is processed by the agent.
Audit Metadata