Author: TechFlow
Crypto assets have warmed up after a period of "Christmas discounts".
Among them, AI Agent-related projects are still the most rebounding, reflecting the continued optimism of capital. However, the differentiation effect in this big track is also very obvious:
Some projects may not come back after falling, while some projects are rebounding in clusters.
Among the projects that have rebounded significantly in recent days, in addition to a few leaders, one type of topic may be worth your attention - using AI Agents to do crypto research and investment.
Don't misunderstand, the current Agents are obviously unable to compete with professional market analysts. Previously, AIXBT, which helped you analyze Alpha projects, became popular;
However, the "research-type" AI agents in this stage are more refined than AIXBT. Some can help you look at K-lines for technical analysis, some can help you analyze whether a project has rug risk, and some can even generate information summaries similar to research reports...
Being able to stand out in a small direction and partially provide references for trading is the reason why these Agents can currently attract attention, after all, they are close to money and trading, and the probability of gaining favor will also be higher.
The TechFlow editorial department has summarized several relatively outstanding projects as follows.
$AGENCY (Agent Scarlett): Your token due diligence agent
CA: 5UmDWgyLV1JBg8Jr8NwyezXdQkiU3vHGJu2efm7Cpump
Current market cap: 12 M
7-day increase: 1100%
Agent Scarlett looks more like a token analysis bot, which can quickly use Scarlett to identify and investigate potential suspicious projects, and fully understand any token.
It is worth mentioning that this AI Agent is also based on the Eliza framework of ai16z, and has had a relatively high discussion on the crypto Twitter in the past few days.
In terms of token economics, 10% of the AGENCY tokens will also be donated to the ai16z DAO.
Functionally, it is currently available for free use by everyone on Telegram and Twitter (usage tutorial).
The usage is also very simple, you can @her on TG or Twitter, and then enter the token contract address you are interested in, and she will automatically return the results to you.
In terms of practicality, one of the better points of Scarlett is that the analysis she returns to you includes the fundamentals and on-chain status of a token, such as telling you whether the social media sentiment and organizations and KOLs related to the token have given support and attention; At the same time, it will also summarize the on-chain data such as the number of holding addresses for you.
At the same time, she will also list out the pros and cons, similar to a short (shallow) research report.
You can also continue to ask her more questions to get more in-depth answers (which also tests your question-asking skills).
However, the editors have seen similar products very early, such as Copilot on Messari and Scopechat on 0x Scope, which are similar to the crypto version of GPT.
But the current market seems to be very receptive to this type of small assistant with tokens, relevant data shows that the discussion heat of $AGENCY is rising rapidly, and the relationship between discussion heat and price can also be clearly seen.
From the perspective of chip distribution, nearly half of the addresses holding less than $100 out of about 3,000 holding addresses, which also reflects the nature of this type of AI Agent being close to retail investors:
Free to use, everyone can play, and if you feel it's not bad, you'll just buy it casually
It's very easy for KOLs to take it out as a target for promotion, after all, an harmless analysis tool is liked by retail investors, and you don't even need to link your wallet
$TRISIG (Tri Sigma 3 σ): Will write research Threads on X
CA: BLDiYcvm3CLcgZ7XUBPgz6idSAkNmWY6MBbm8Xpjpump
Current market cap: 33 M
7-day increase: 578%
TriSig is a smart crypto analyst who claims to be able to identify early alpha projects, and answer challenging questions in a simplified way, and share his views on any major events or trends in the crypto market.
Its name and image are more mathematically oriented (corresponding to the mathematical symbol σ: sigma), leaning more towards the omniscient AI creator.
Functionally, it's similar to the aforementioned $Agency, just ask questions about tokens and it will answer.
But the difference is that the TriSig on TG is currently just an entertainment bot. Serious analysis dialogues are limited to X.
You just need to @tri_sigma_ on X to use it for analysis in any tweet interaction.
Interestingly, it will write tweets itself, doing a quick research introduction to a project, including what a project is doing, current social media sentiment analysis, and its own subjective bullish or bearish judgment.
In addition, this project is also very community-oriented. If TriSig is deeply impressed by the question you asked, it will airdrop $TriSig tokens by asking you to leave your wallet address in the tweet reply.
Further data shows that the TRISIG token price has risen rapidly in the past 2-3 days, corresponding to the gradual increase in external discussion; as of today, the price has seen a significant pullback, but the discussion has not decreased.
In terms of chip structure, the difference between addresses holding less than $100 and less than $1,000 is not large, compared to the previous AGENCY project, this token is relatively more favored by those who hold large positions, but its market cap is also higher.
$KWANT (kwantxbt): Focus on technical analysis, provide price levels
CA: 9Yt5tHLFB2Uz1yg3cyEpTN4KTSWhiGpKxXPJ8HX3hat
Current market cap: 6 M
7-day increase: 25% (24h 70% rebound)
If the previous two projects still have fundamental research, Kwant is more direct, directly analyzing the token's price chart and providing investment recommendations.
The usage is also similar, just @KwantAI_bot in TG and send the token's CA, and you can get pure technical analysis, including but not limited to the relationship between volume and price, chart patterns, and subsequent operation suggestions.
Unlike general projects that stack up and say "NFA" (not financial advice), KWANT directly gives you very specific operational suggestions such as support levels, breakout levels and stop loss levels.
Of course, any crypto investment has huge volatility and risk, and KWANT's suggestions can't be invincible either; but when the AI Agent has more information and has the ability to continuously learn and evolve, would you believe the entry points given by the "group big guys", or believe the AI?
Having one more reference and voice is not a bad thing after all, this is the bottom line for these projects to be sought after.
In addition, since the KWANT token has been released for more than half a month, the recent social media sentiment is not very meaningful as a reference, because there will always be someone discussing KWANT after a while, just looking at the price, the recent rebound trend of the token is very obvious.
In terms of chip structure, small retail investors still constitute the basic pool, and users with holdings of more than 25K are not many. The author believes that one possible reason is:
Compared to bots that research the fundamentals, this type of direct point-giving has a higher rate of derailment, and it is very easy to evolve into a negative teaching material that is often wrong.
$COOKIE (Cookie.fun): Providing data and sentiment analysis for all AI coins
CA: 0xc0041ef357b183448b235a8ea73ce4e4ec8c265f (available on both BSC and BASE)
Current market cap: 24M
7-day increase: 7%
Note that this project is not an AI Agent, but a data analysis platform that has risen rapidly in the AI Agent craze.
The reason it is also listed for review is that the editor and analysts on the external network also use this platform to observe the current market's hot trends, which can provide a lot of meaningful data for investment research;
And this platform also has its own token $COOKIE, one of the important functions of which is that you can stake it to use more data analysis functions of the platform, such as viewing the on-chain chip structure, whale holdings, and more decision-making assistance sentiment indicators of a certain AI Agent token.
In the attention-driven cryptocurrency market, understanding the correlation between sentiment and price is crucial for analyzing a project, so platforms like Cookie that focus on data analysis of AI Agents naturally have their raison d'être, and their focus on AI Agent data analysis can also quickly occupy an ecological position in the craze.
In contrast, the platform token is not a MEME, and in a strict sense it does not belong to the AI Agent coin, so the price fluctuation range is also limited, and it is more about providing practical value.
Where are we?
From joking MEME to trying to guide you to "make money", the AI Agents in the cryptocurrency market are constantly evolving, and within a few months, the business model has fully blossomed.
Whether the result is good or not remains to be seen, but this full bloom does have a certain feeling of the DeFi Summer at the time.
If you look back at the timeline of the last few months of 2024, you can clearly see a thread:
From having no real meaning, to trying to find more real meaning, AI Agent tokens are seeking more functions to support the narrative.
For example, social, music making, helping you build development frameworks or investment research, the current market seems to feel ashamed to say its selling point if it is not related to AI.
The only thing we can do is to clearly identify which stage we are in, and make corresponding operations and judgments based on the stage's hot spots.
As for when the clamor will end and the final chapter will come, it is a question that cannot be foreseen and worrying about it is useless. After all, there are always people who pay the bill in the feast, and there are always people who benefit from the trend.
Grasping the trend is the main melody of the cryptocurrency market.
