Glossary
A complete reference of domain terms used throughout the Chainabit platform.
Core Domain
Chainer
The individual (or AI agent) using the system. A Chainer is a person who actively builds their future through consistent, verifiable action. Every authenticated user on the platform is a Chainer.
Chainy
A goal-scoped micro-workspace. A Chainy contains execution protocols (Chains), direct tasks (Bits), a kanban board, and behavioral visualizations for a specific project or vision. Think of it as a GitHub repository — the organizing container for all work toward a goal.
Chain
A behavioral execution ledger: a cryptographically signed, time-structured record of a Chainer's declared commitment and actual execution history within a Chainy. Like a signed Git branch with a time-boxed commit policy, a Chain produces attestations (period signatures) and milestone certificates that any third party can independently verify as proof of effort.
Bit
The smallest executable unit of progress. A Bit has dual identity: as a pending task (like a GitHub issue) and, upon completion, as a commit — an immutable execution event recorded in bit_completion_events with a hash-linked previous event reference.
Protocol Type
The behavioral classification of a Chain, defining its execution semantics. Six types: scheduled (recurring), constraint (abstention), event_driven, ai_triggered, adaptive (AI adjusts cadence), and one_shot.
Verification Level
The trust tier of a bit completion, from self_attested (user clicked checkbox) through ai_verified (AI-scored) to cryptographic (IDE/CI-backed). Chains define a minimum required_verification_level.
Chain Period
A time-boxed evaluation window within a Chain (e.g., one day, one week). Defines required_count and completed_count. When requirements are met, a period signature is created.
Chain Period Signature
A cryptographic attestation that a period's execution requirements were met. Contains a SHA256 signature_hash, AI verification score, and optional AI attestation data. Signatures form the proof layer of the behavioral ledger.
Certificate
A milestone proof issued when a Chain reaches achievement thresholds (completion history milestones, completion counts, consistency rates). Certificates are polymorphic, NFT-mintable, and exportable as W3C Verifiable Credentials.
Attestation
The act of cryptographically signing a chain period's execution results. Attestations can come from users, AI (as verifiers or co-executors), or the system.
Chao
The AI advisor trained on system continuity and behavioral patterns. Chao provides personalized guidance based on a Chainer's execution history.
Organization
Account
An enterprise-level organization that owns workspaces. Accounts serve as the top-level container for billing, team management, and resource ownership. A single account can contain multiple workspaces.
Workspace
A scoped environment within an account for organizing agents and workflows. Workspaces provide logical separation of resources, enabling teams to manage different projects or contexts independently under the same account.
Gamification
Duo Race
A competitive bit-completion race between two teams. Each bit completed during an active race scores points for the performer's team.
Season
A time-bounded competition window for seasonal gamification events, challenges, and leaderboard evaluation.
Badge
A criteria-driven achievement that chainers can earn. Badges have rarity tiers (common through legendary) and grant XP rewards.
Leaderboard
Ranked scoring of chainers by performance metrics (run history, completions, consistency) within seasons or all-time.
Billing & Access
Entitlement
A feature gate that determines what capabilities a user has based on their subscription. Entitlements control access to specific platform features, API endpoints, and usage limits.
Credit
The unit of consumption for AI features, tracked via a wallet system. Credits are deducted when using AI capabilities such as running agents, executing workflows, or interacting with Twins.
AI & Automation
Agent
An AI entity configured with tools, scopes, and behavioral rules. Agents can be assigned to workspaces and given specific skills to perform tasks autonomously.
Twin
An AI behavioral identity that persists personality and memory across interactions. A Twin is bound to a specific Chainer and accumulates context over time. The Twin stores structured identity facets (goals, style profile, tool preferences, expertise signals) and evolves after each chain/agent completion.
Twin Evolution
The process by which a Twin's identity facets are updated based on chain execution results, agent run outcomes, and user corrections. Style inference extracts tone/format preferences; goal inference extracts objectives from chain topics.
Context
A curated knowledge bundle that Chao reasons over. A Context groups multiple sources (documents, connector snapshots, chain outputs, web pages, notes) into a named collection. Contexts can be bound to chains and shared across workspaces.
Context Source
An individual knowledge source within a Context. Source types include document (uploaded files), connector_snapshot (live data from connected tools), chain_output (results of chain executions), web (fetched URLs), and note (free-text).
Agent Run
A goal-directed autonomous execution where an AI Agent decomposes a high-level objective into sub-tasks, executes each step, evaluates results, and re-plans until the goal is complete. Agent Runs include budget enforcement and approval gates for risky actions.
Chainy Instruction
A custom AI instruction set per Chainy. When present, the instruction is injected into the Chao prompt only at the start of a session (first message), not on every turn.
AI Directive
An AI-generated recommendation for chain adaptation: schedule adjustments, difficulty changes, bit decomposition, or certificate triggers. Directives can be applied or dismissed by the Chainer.
Skill
A versioned capability that can be attached to agents. Skills define what an agent can do — from calling external APIs to processing data.
Workflow
A DAG-based automation pipeline connecting multiple agent actions. Workflows define a directed acyclic graph of steps for complex multi-step automations.
Run
A single execution of an AI feature, agent, or workflow. A Run transitions through defined states (queued, processing, completed, failed) and can be monitored in real time via Server-Sent Events.
Session
A conversational context for AI interactions. Sessions maintain message history and state across multiple exchanges.