Author: Cambrian Network

Compiled by: Deep Tide TechFlow

Deep Dive: This is the most comprehensive quarterly industry report to date on the "AI Agent + Encrypted Infrastructure" convergence track, published by Cambrian Network. The report goes beyond simply listing key information; it analyzes the evolution of the entire market: from rule-based decision-making to intelligent LLM, from user-held keys to autonomous agent execution, and the latest implementation progress of key infrastructure such as the x402 payment standard and the ERC-8004 identity protocol. This report is essential reading for understanding the current landscape and future prospects of the AgentFi track by 2026.

The full text is as follows:

Since the release of our last Agentic Finance Landscape Report in December 2025, this market segment has experienced explosive growth. We've seen a constant stream of new projects, rising adoption rates, and autonomous agents operating with the help of crypto infrastructure. The data speaks for itself: x402 payments continue to accelerate, completing over 15 million transactions in the past 30 days and establishing institutional partnerships with major companies like AWS and Stripe. ERC-8004, a standard designed to build trust between agents and the marketplace, launched on the mainnet in late January.

Agentic Finance Panorama Report Q1 2026

This report surveys a comprehensive overview of autonomous products that support retail users' participation in decentralized finance (DeFi). As before, we have established strict selection criteria: projects must be active and publicly available, with real users and operational funds. Agents are categorized by product type.

Note: In this report, products previously classified as "Liquidity Provider Agent" and "Lending Agent" have been merged into the broader "Rewards" category, reflecting the increasing diversification of Agents' ways of earning rewards.

What is Agentic Finance?

Agentic Finance is an emerging market segment where products use automated technologies to proactively manage user funds or provide financial advice. Inspired by ChatGPT, Claude Code, and emerging open-source agent tools such as OpenClaw and Moltbot, some products use LLM, while others employ hard-coded rules or traditional machine learning. Regardless of the underlying technology, many of them describe themselves as "agentic."

Early forms of agentic finance were simple automated trading bots. The DeFi summer of 2020-2021 brought yield farming bots and Vault protocols like Yearn that could automatically harvest and compound interest, but the logic remained deterministic: if X holds, then Y is executed. By 2022-2023, automation had expanded across protocols and chains (bots manage collateral ratios, rebalance LP positions, and chase yields in lending markets), but users still held the keys and set every parameter.

The first real shift occurred in 2024 when LLMs entered the technology stack, and we began to see natural language interfaces, AI-driven analytics agents, and conversational trading. The decisive breakthrough came in 2025—the first closed-loop agents that acted according to strict rules and required no human approval emerged. Today, Agentic Finance is a recognized market category, and the Cambrian team is proud to have pioneered the use of the terms "agentic finance" and "AgentFi" to describe this segment. The products covered in this report represent the next major step in the development of AgentFi: software capable of making decisions, managing capital, and paying for its own operations.

Autonomy and intelligence of Agentic Finance

The Agentic Finance project spans multiple professional use cases, but each project can be placed on a simple coordinate system. The horizontal axis represents the level of intelligence: the left side represents rule-based systems that rely on mathematics and statistics, and the right side represents LLM-based systems (which many now call "AI Agents"). The vertical axis represents the level of autonomy: the bottom represents tools that provide information, the middle represents systems that require human intervention, and the top represents systems that execute autonomously within the scope of a policy and do not require approval.

Since the first issue of this report was published, the market focus has shifted upwards, moving towards greater autonomy. However, the largest asset management products still tend to rely on rules-based approaches in capital allocation decisions. While LLM (Local Level Managers) have demonstrated their advantages in areas such as interface, information retrieval, and analysis, algorithm-based rule-based agents remain the mainstream when it comes to fund flows, as reliability and auditability are paramount.

AgentFi Market Growth in 2025-2026

Growth data

Since the last report was published, the ecosystem has continued to expand. The number of unique users of on-chain AgentFi products has continued to climb. In the past 30 days, Bankr and Giza have led in user deposits, each with 900+ deposits.

Among the leading agent projects in this field, profit-seeking agents are at the center of discussion alongside trading and portfolio optimization agents.

We are also seeing increasing institutional interest and active investment in building the infrastructure needed for the exponential growth of Agentic Finance. Examples include Grayscale's launch of a decentralized AI fund in late 2025, Robinhood's release of the Cortex Assistant for retail users, and Visa's Trusted Agent Protocol, based on the Coinbase x402 payment standard, for AI-driven business scenarios.

x402 continues to expand on new chains, now supporting BSC and Polygon in addition to the already supported Base and Solana chains. Despite a slowdown due to recent market conditions, agent payment volume remains strong, with x402's cumulative transaction volume exceeding $50 million. Over the past 30 days, the majority of x402 transaction volume has been processed by Virtuals.

Emerging Trends

If you're active on X, you may have noticed that the timeline has recently been dominated by OpenClaw-style agents built on the OpenClaw core framework, such as clawdbotatg. These agents are capable of autonomously completing payments, minting tokens, executing AI-driven crypto transactions, and participating in prediction markets.

Bankr has become a widely used execution layer for these OpenClaw Agents, providing programmable wallets and automatic fee redirection to help cover operating costs. And it's not just Bankr—currently, over 80 projects are building Agentic infrastructure on Base for OpenClaw-style agents, enabling them to participate in Agentic Finance.

Various emerging agents are also actively utilizing the recently launched ERC-8004 to register their identities as NFTs and build a reputation history to prove their legitimacy. Since ERC-8004 went live on January 29, more than 24,000 agents have completed registration.

With all the standards mentioned in the previous report (ERC-8004, x402, AP2, ACP) combined, crypto infrastructure is proving itself to be the most powerful underlying layer for AI agents to hold, transfer, and create monetary value, without any gatekeepers. A recent example is Coinbase's newly launched Agentic Wallets.

Agentic Finance Overview Q1 2026

In early 2026, retail users were early adopters of Agentic Finance, while institutions were paying attention but full adoption was still a long way off. Below is a list of the most representative AgentFi products for retail.

Cambrian has strict selection criteria for projects included in the Agentic Finance panorama: we do not include projects that are still under development or in internal testing, nor do we include projects that only provide an LLM interface but require humans to make "critical" decisions. This excludes many projects.

Trading and Portfolio Optimization Agent

Trading agents are the first products that come to mind for most people when they hear "agentic finance." These agents manage users' funds by rebalancing portfolios or selecting assets to buy and sell. Trading decisions require access to exchanges, tradable assets, budgets, trading guidelines or rules, and high-quality data. The following tools support one or more of these aspects.

@askjimmy_ai: AI signal terminal with win rate/loss/profit indicators, supports one-click execution.

@HeyAnonai: Spot and leveraged trading configured with natural language prompts

@get_truenorth: Trading functionality powered by the advanced discovery and research engine.

@AIWayfinder: Utilizing on-chain agents for autonomous cross-chain transactions

@bankrbot: A versatile DeFi terminal that supports cross-chain exchanges and automated strategies.

@glider_fi: Automated, customizable portfolio management

@AgentHustleAI: DeFi personal assistant, covering everything from basic exchanges to custom automation strategies.

@Velvet_Capital: A DeFi application equipped with an information assistant that can perform some basic operations (such as exchanging).

@SurfAI: DeFi Research and Execution Assistant combining in-depth market analysis and automated trading workflows

@HeyElsaAI: An AI assistant for managing crypto assets, executing DeFi transactions, and providing real-time cross-chain data.

@elfa_ai: An analytics-driven trading assistant that integrates social signals and smart wallet tracking.

@ethy_agent: Autonomous trading assistant that executes transactions, staking, yielding, and transfers directly from the user's smart wallet.

@symphonyio: A cross-chain DeFi execution terminal with AI-assisted trading capabilities.

@Cod3xOrg: An event-driven trading engine equipped with a terminal that supports updatable algorithmic trading strategies.

@autonolas: Supports adaptive DeFi strategies through a self-contained portfolio management agent deployed on a desktop.

@Butler_Agent: Orchestrate agents to perform transactions, yield farming, and DeFi automation via Virtuals Protocol Agent.

@fere_ai: An AI assistant that supports cross-chain, Meme coin, prediction market, and DeFi protocol trading.

@modenetwork: A perpetual DEX with an LLM-driven trading agent that executes 24/7 based on user-configured strategies.

@minara: An AI trading assistant with perpetual contract support and customizable agent workflows

@MiloOnChains: Solana trading agent, supports natural language execution and optional 24-hour automated trading.

Revenue Agent

Earning yield on AgentFi takes many forms, with the most popular currently being through lending protocols (borrowers pay interest, lenders receive returns). Users can also provide liquidity in DEX (decentralized exchange) liquidity pools to earn trading fees, or deposit funds into select Vaults that allocate funds across multiple venues. The following agents help users find the best risk-adjusted returns among these strategies.

Note: In previous reports, we categorized LP Agents and Revenue Agents as separate entities. Over time, we have observed an increasing number of projects integrating multiple revenue generation strategies, including but not limited to lending, LPs, and spot trading.

@afiprotocol_xyz: Achieving yield optimization through non-custodial algorithmic agents managing liquidity and derivatives strategies.

@almanak: Multi-strategy return optimization using AI-generated strategies

ARMA by @gizatechxyz: Stablecoin Yield Agent

@ArrakisFinance: Automated Market Making Strategy

@getaxal: Finding the optimal lending protocols, liquidity pools, and Delta-neutral yield strategies

@DeFiSaver: Lending across all major protocols, providing automatic anti-liquidation protection.

@Infinit_Labs: A DeFi abstraction layer that provides one-click agent presets and suggestive DeFi strategies.

@kamino: An automated lending protocol on Solana that offers optimized returns, leveraged positions, and modular credit markets.

@uselulo: Generate returns by depositing funds into an over-collateralized lending pool.

@mamo: A financial assistant designed to simplify personal financial management

@pendle_fi: Autonomous return optimization across lending protocols, Vault, and Delta neutral strategies

@SaildotMoney: Automatically rotate funds between lending pools, Vaults, and AMMs.

@superformxyz: A non-custodial on-chain digital bank that aggregates yields from cross-chain lending protocols and supports cross-chain deposits.

@Surf_Liquid: An on-chain DeFi agent that autonomously manages stablecoin yields and has a verifiable rebalancing record.

@RFLnow: An autonomous agent on the base, managing LP yield farming end-to-end, including deploying liquidity, harvesting, rebalancing, and optimizing returns.

@Zyfai_: Loan Income Management

Prediction and Game Theory Agent

Prediction markets are platforms where users can trade on the outcomes of future events, such as election results or sporting events. These markets typically require tracking news and other real-time, changing real-world information. Prediction markets are one of the most exciting emerging categories in AgentFi. Agents' ability to acquire and track real-world event information from a wider range of data sources than humans can manually process makes them ideal participants in prediction markets.

We expect this segment to continue growing through 2026. Academic research (such as this study by the Federal Reserve) suggests that prediction markets can provide valuable insights for policymakers. Meanwhile, the CFTC is asserting its authority to legitimize prediction markets, overriding state opposition.

@AskBillyBets: Active in the crypto-based sports betting and prediction market

@sire_agent: An AI-driven autonomous sports betting fund

Sentiment, Fundamentals, News and Technical Analysis Agent

Investors typically use market analysis to determine what to buy and sentiment analysis to decide when to buy and sell. LLM has significantly changed market and sentiment analysis by expanding the amount of data analyzed, accelerating the analysis process, and forming a deeper contextual understanding by identifying connections between data sources. A key difference between analytics agents and the aforementioned agents is that they do not take direct action but rather provide informative guidance. There are numerous analytics agents; only a few are listed below.

@aixbt_agent: Crypto market intelligence, narrative detection, alpha analysis, KOL tracking

@DeepFortyTwo: Alpha detection agent for high-signal tweets on X, scoring and analyzing on-chain and off-chain information.

LlamaAI by @DefiLlama: A DeFi data assistant that queries the DefiLlama database to obtain TVL, yield, fees, and token analysis.

Messari Copilot by @MessariCrypto: Providing citation-supported answers to questions driven by Messari research and real-time data.

What's next for Agentic Finance in 2026?

The first quarter of 2026 has been a period of stress testing for crypto infrastructure. The total crypto market capitalization has retreated somewhat amid macroeconomic uncertainty after surpassing a peak of $4 trillion. However, BlackRock's IBIT (iShares Bitcoin Trust) recorded a historic single-day trading volume of $10 billion in early February; tokenized RWA has exceeded $20 billion (with tokenized Treasury bonds alone reaching $8-10 billion, up from $5.5 billion in the third quarter of 2025); and the stablecoin supply has climbed to $310 billion.

Regulatory clarity and its integration with DeFi, TradeFi, and AI

The regulatory landscape continues to evolve towards providing clear rules for the sustained growth of the crypto economy. We believe this is a significant step forward, as regulatory clarity is key to bringing more capital into the crypto space for TradeFi. For example, the SEC, under Chairman Paul Atkins, has announced "innovation exemptions" and upcoming "crypto regulation" rules, while the CLARITY Act (which clarifies CFTC/SEC jurisdiction) has a realistic possibility of passing before the November 2026 midterm elections. Meanwhile, Hong Kong is preparing to issue its first stablecoin licenses in March 2026; the UK's FCA (Financial Conduct Authority) has opened a regulatory sandbox supporting the issuance of stablecoins in the UK; and MiCA's full CASP (Crypto Asset Service Provider) compliance deadline is July 2026—Circle's EURC has become the dominant compliant euro stablecoin after several EU exchanges delisted USDT last year.

Regulatory clarity, combined with proven crypto infrastructure, is laying the foundation for on-chain institutional capital and the scaling up of agentic finance. In the coming months, we expect both institutional adoption and agentic finance to continue growing, but along parallel paths. Agentic finance must first establish a credible track record with retail users before institutions will allocate meaningful funds; only then will these two paths converge.

We also anticipate continued growth in stablecoin usage and agent payments as new autonomous agent trends (such as OpenClaw) emerge and require programmatic access to APIs, computing power, and third-party services. As a native settlement layer for machine-to-machine transactions, stablecoins are poised to become a key catalyst for the mainstreaming of agentic finance.