Every AI, email, and agent term explained for developers. From context engineering to DKIM.
Filter by category
Google's open protocol for AI agents to discover, communicate, and collaborate with each other across different platforms.
An open protocol that standardizes how AI agents discover each other and exchange structured messages.
The verifiable identity assigned to an AI agent, enabling it to authenticate, send email, and interact with external services as a distinct entity.
A large group of AI agents operating in parallel with minimal central coordination, often self-organizing around a shared objective.
AI systems that can autonomously plan, make decisions, and take actions to accomplish goals with minimal human intervention.
Low-quality, mass-produced AI-generated content that adds noise without providing genuine value.
A neural network component that lets a model dynamically focus on the most relevant parts of its input when producing each part of the output.
Standardized tests used to measure and compare AI model performance across specific tasks like reasoning, coding, math, and language understanding.
A prompting technique that instructs an LLM to reason through a problem step by step before producing a final answer.
A marketplace for AI agent skills where agents can discover, install, and use new capabilities without human configuration.
The discipline of designing and managing everything that flows into an AI model's context window — not just the prompt, but instructions, state, tool definitions, memory, and output constraints.
The maximum amount of text a language model can process in a single request, measured in tokens.
A holding area for messages that couldn't be processed or delivered after exhausting all retry attempts.
A technique where a smaller 'student' model is trained to replicate the behavior of a larger 'teacher' model, producing a compact model that retains much of the original's capability.
A DNS-based email authentication method that lets receiving servers verify a message was actually sent and unmodified by the domain it claims to come from.
An email policy protocol that tells receiving servers what to do when SPF or DKIM checks fail for messages claiming to be from your domain.
The requirement that the domain in SPF or DKIM authentication matches the domain in the email's visible From header.
Numerical vector representations of text, images, or other data that capture semantic meaning, enabling similarity search and machine learning tasks.
The return-path address used during SMTP transmission, which may differ from the visible From address the recipient sees.
A memory system that stores and retrieves specific past experiences or interactions, giving an AI agent the ability to recall what happened in previous sessions.
A retry strategy where the wait time between attempts increases exponentially, reducing load on failing systems.
The process of further training a pre-trained AI model on a specific dataset to specialize its behavior for a particular task.
An anti-spam technique where a mail server temporarily rejects emails from unknown senders, expecting legitimate servers to retry.
Constraints and safety mechanisms built around AI agents to prevent harmful, off-topic, or unauthorized behavior.
When an AI model generates confident-sounding information that is factually incorrect or entirely fabricated.
A permanent email delivery failure caused by an invalid address, non-existent domain, or blocked recipient.
A cryptographic hash used to verify that a webhook payload was sent by a trusted source and hasn't been tampered with.
A design pattern where an AI agent pauses its workflow to get human approval, review, or input before proceeding with a critical action.
A property where performing the same operation multiple times produces the same result as performing it once.
The automated process of creating and configuring a dedicated email inbox for an AI agent, including address assignment, DNS setup, and access credentials.
The process of running input data through a trained AI model to get a prediction or output. Every API call to an LLM is an inference request.
The practice of gradually increasing email volume from a new IP address to build sender reputation with email providers.
A security principle where an agent is granted only the minimum permissions it needs to perform its task, and nothing more.
Low-Rank Adaptation — a fine-tuning technique that trains a small set of additional parameters instead of modifying the entire model, making customization fast and memory-efficient.
An open protocol created by Anthropic that standardizes how AI models connect to external tools, data sources, and services through a universal interface.
A model architecture that uses a routing mechanism to activate only a subset of specialized sub-networks (experts) for each input, increasing capacity without proportionally increasing compute.
Architectures where multiple AI agents work together, each handling specialized tasks, to accomplish goals that a single agent cannot.
An architecture where a single system serves multiple independent customers or agents with isolated data, access, and configuration.
A DNS record that specifies which mail server should receive email for a domain.
A central AI agent that coordinates, delegates, and manages tasks across multiple sub-agents in a multi-agent system.
An attack where malicious input tricks an AI model into ignoring its instructions and performing unintended actions.
A technique that reduces the precision of a model's numerical weights to shrink its size and speed up inference, with minimal loss in quality.
A technique that improves AI responses by retrieving relevant documents and including them as context before generating an answer.
An agent architecture where the model alternates between reasoning about a situation and taking actions, using observations from each action to inform the next step.
Reinforcement Learning from Human Feedback — a training technique where human preferences are used to fine-tune an AI model's behavior, making it more helpful, harmless, and honest.
An isolated email inbox where an agent can only access its own messages and cannot read, modify, or interfere with other agents' inboxes.
A score that email providers assign to a sending domain or IP address based on its email-sending behavior and history.
A markdown file that describes an agent skill's capabilities, inputs, and usage instructions so other agents and platforms can discover and use it.
A server that forwards outbound email on behalf of another system, handling delivery, authentication, and retry logic.
Three-digit status codes returned by mail servers during SMTP transactions to indicate success, temporary failure, or permanent rejection.
A temporary email delivery failure where the message may succeed if retried, typically caused by a full mailbox or server issue.
A DNS record that lists which mail servers are authorized to send email on behalf of your domain.
An AI agent that does not retain memory between interactions, treating each request as independent with no prior context.
AI model responses formatted in a predictable schema like JSON, enabling reliable machine-to-machine communication.
A list of email addresses that should never receive messages, typically populated by hard bounces, unsubscribes, and spam complaints.
A hidden instruction given to an LLM before the user's message that defines the model's role, behavior, constraints, and personality.
A parameter that controls the randomness of an LLM's output, where lower values produce more predictable responses and higher values produce more creative ones.
The basic units of text that language models read and generate, roughly equivalent to three-quarters of a word.
The ability of an AI model to call external functions, APIs, or services to take actions and retrieve real-time information.
The neural network architecture behind all modern LLMs, using self-attention to process sequences of tokens in parallel rather than one at a time.
A database optimized for storing, indexing, and querying high-dimensional vectors, commonly used to power semantic search and RAG in AI applications.
A development approach where programmers describe what they want in natural language and let AI generate the code.
An HTTP callback that sends real-time event data to your application when something happens, like an email being delivered or bouncing.