> For the complete documentation index, see [llms.txt](https://blockgram.gitbook.io/blockgram-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://blockgram.gitbook.io/blockgram-docs/about-blockgram/core-features/have-ai-assistant-right-beside-your-keyboard.md). # Have AI Assistant Right Beside Your Keyboard BlockGram brings real-time, context-aware artificial intelligence directly into Telegram chats, turning each conversation into an intelligent interface. The AI assistant is embedded in the chat environment, offering actionable suggestions, parsing commands, summarizing discussions, generating replies, and enabling crypto interactions, all without leaving the messaging interface. #### Core Capabilities of the AI Assistant **Inline Prompting and Completion** * Suggests auto-completions for commands like “send 0.5 MATIC to…” * Summarizes recent messages on demand * Detects intents for payments, bill splitting, or portfolio updates and prompts users proactively **Proactive Context Analysis** * Continuously monitors conversation for relevant triggers * Detects mentions of debts, expenses, or transactions * Suggests initiating payments or splitting bills * Recognizes wallet addresses and offers secure sending options **Multimodal Interaction** * Supports typed natural language in major languages * Voice command recognition via speech-to-text integration * Structured quick actions for recent transactions, NFTs, or assets #### Technical Architecture **Inference Pipeline** * **Input Parsing:** User messages are tokenized and semantically embedded using lightweight transformer models * **Intent Recognition:** Classifies messages into categories such as payment, request, summary, schedule, or crypto transfer * **Response Composition:** Crafts context-aware replies or prepares actions based on detected intent * **Action Dispatch:** Executes AI-suggested actions via smart contracts or integrated APIs #### Use Cases Within Chat * Instantly generate wallets or send crypto based on natural commands * Summarize group chats or DAO discussions in real time * Automate recurring payments or shared expense calculations * Provide portfolio insights or NFT ownership confirmations directly in chat #### Customization and Control * **Privacy-Aware Context Windows:** Only recent relevant messages are processed; data isn’t permanently stored * **On-Demand Invocation:** Users choose whether AI acts automatically, via keyword triggers, or through manual slash commands * **Model Switching:** Advanced users can select lightweight local models or high-power remote inference #### Privacy and Safety * **On-Device Inference:** Reduces latency and keeps sensitive data local when hardware supports it * **No Data Retention:** AI inputs are ephemeral unless encrypted and stored by the user * **AI Explanation Layer:** Before executing any action, users see what the AI interpreted and what it plans to do, allowing full manual confirmation #### Benefits * Integrates financial intelligence and conversational support directly into Telegram * Reduces cognitive load and manual steps for crypto tasks * Provides a personalized assistant for everyday chat and financial workflows * Bridges natural language commands with blockchain operations, making crypto accessible without technical knowledge --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://blockgram.gitbook.io/blockgram-docs/about-blockgram/core-features/have-ai-assistant-right-beside-your-keyboard.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.