DEV Community

gxqwb
gxqwb

Posted on

Two Solana Sub-$1M Projects That Survive a Real Screen — and Why Most Don't

#solana #defi #crypto #analysis

Mycelium pairs $MYC liquidity directly with selected Solana tokens. The obvious question for anyone watching that strategy: what else belongs in the network? So I ran the filters the brief specifies against the whole Solana market, on-chain, and published what came out — including the part where almost nothing does.

Two projects qualify. One more passes every number but fails the paper trail, and I explain why it is not presented as a pick.

The screen

Every candidate had to clear all of these, measured on 2026-10-25:

  • On Solana
  • Circulating market cap below $1M at research time
  • A product people can use today
  • Not a meme token (playful branding allowed if the product is substantive)
  • Trading for at least 60 days
  • At least $10,000 of 24-hour trading volume
  • At least $50,000 of current DEX liquidity across verified pools, counting each pool once
  • Evidence of a product update, development activity, or substantive team communication in the last 30 days
  • Not already in the Mycelium Strategy — $KNOTS, $PRCL, $GABLE, $KINS, $DREAMS, $GUAC, $STREAM, $ESF
  • Not $SOL, $USDC, $USDT, $MYC or $ESF

Market cap and liquidity are from DexScreener; holder counts and top-10 concentration are from Jupiter's token audit. Supply, mint authority and freeze authority are read directly from the mint account on Solana mainnet.

Method, and the shape of the problem

Jupiter's verified-token list held 4,042 Solana tokens. Of those, 68 cleared market cap below $1M, at least $50K of liquidity and at least $10K of 24-hour volume.

Then the 68 had to be classified by hand, because the sub-$1M band on Solana is not what it looks like:

  • 16 are tokenized equities issued under Backpack Securities — NVDA, AAPL, TSLA, AMZN, MSTR, NFLX, HOOD, LLY, DKNG, NKE, BABA, HIMS, AMD and others. Real products, but not Solana projects.
  • 4 are wrapped or bridged majors — BNB (Portal), cbLTC, AUDIO (Portal), and a Wormhole-wrapped meme.
  • Most of the rest are memes, thin wrappers of large-cap tokens, or index products.
  • A handful are genuine native projects. Two of them pass everything.

That distribution is the finding. Below $1M, liquidity is now concentrated in tokenized stocks rather than in small native protocols.

Pick 1 — AgenC ($AgenC)

Product. An autonomous agent marketplace on Solana mainnet: post a task, an agent delivers, review settles the reward on-chain, and every payout lands as a receipt. The site reports protocol R5 live, core version 0.19.13, and streams settlement instructions as they happen.

Evidence it meets the filters:

  • Solana mainnet, token mint 5yC9BM8KUsJTPbWPLfA2N8qH1s9V8DQ3Vcw1G6Jdpump
  • Market cap $806,040, FDV identical — DexScreener, 2026-10-25
  • Working product live at agenc.tech with an active on-chain activity feed
  • Not a meme: a settlement protocol with published docs, SDK and REST API
  • First pool created 2026-02-24 — 243 days of trading
  • 24-hour volume $51,629
  • Liquidity $148,878 across 16 distinct pools; $139,961 in SOL/USDC-quoted pools, led by a PumpSwap AgenC/SOL pool at $131,729
  • Activity in the last 30 days: news dated Oct 18 (AgenC Studios), Oct 16 (76.2% on Terminal-Bench 2.1 for $34), Oct 14 (core 0.19); the public repo tetsuo-ai/agenc-protocol last pushed Oct 21
  • Not in the Mycelium Strategy

Concentration. Supply is 999,931,062 with no mint authority and no freeze authority — supply cannot be inflated and holders cannot be frozen. 4,576 holders; top-10 wallets hold 54.4% of supply, the weakest number in this analysis. Pools hold roughly 92.4M tokens, about 9.2% of supply, so the concentration sits with wallets rather than with liquidity.

Liquidity safety. Selling $100 costs 0.00% in price impact, $500 costs 0.52%, $2,000 costs 2.49%. For a sub-$1M token that is unusually good — it reflects $131K sitting in one pool rather than spread thin.

Risk. 54.4% in ten wallets is the whole thesis risk. A single large holder exiting would move the price far more than the pool depth suggests. The product is also early: agent marketplaces have strong demos and weak retention, and revenue depends on buyers actually posting paid tasks.

Pick 2 — three.ws ($three)

Product. A 3D agent layer: make an AI agent and 3D avatar from a selfie or a prompt, forge textured 3D models, embed the agent on any site, browse registered agents, and buy or sell agent skills. Also ships a launchpad and a live feed of agent wallet activity.

Evidence it meets the filters:

  • Solana, mint FeMbDoX7R1Psc4GEcvJdsbNbZA3bfztcyDCatJVJpump
  • Market cap $294,394 — DexScreener, 2026-10-25
  • Working product live at three.ws with agent creation, forge, marketplace and console
  • Not a meme: a developer platform with an MCP tooling layer
  • First pool created 2026-04-29 — 179 days of trading
  • 24-hour volume $11,915
  • Liquidity $121,464 across 13 distinct pools; $121,106 in SOL/USDC-quoted pools, led by a PumpSwap three/SOL pool at $115,384
  • Activity in the last 30 days: the public repo nirholas/three.ws (249 stars) was pushed 2026-10-25, with commits adding an /api/mcp-tools endpoint and refreshing feature docs
  • Not in the Mycelium Strategy

Concentration. 14,995 holders and top-10 wallets at 13.7% — the flattest distribution of every candidate that passed the numbers. Pools hold roughly 206.2M tokens, about 20.6% of supply, so a fifth of the supply is already market-making depth rather than overhang.

Liquidity safety. $100 costs 1.49%, $500 costs 2.17%, $2,000 costs 4.04%. Thinner than AgenC but still workable at retail size.

Risk. Volume at $11,915 is barely over the threshold — this is a token that trades, but not much. Low float in pools plus low volume means the price is easy to move in both directions, and 20.6% of supply sitting in pools is depth only for as long as liquidity providers stay.

What passed the numbers but is not a pick

perpspad ($PERPSPAD) — a Solana launchpad where every coin is backed by a live perpetual on Phoenix and Meteora, with mint and freeze authority revoked at creation. Market cap $725,569, liquidity $136,624, 24-hour volume $43,272, 2,730 holders, top-10 24.9%, $2,000 exit slippage 3.94%. It clears every quantitative filter. It fails the last one: I could find no dated product update, no public repository and no dated team communication inside 30 days. That is a documentation gap, not proof of inactivity — but the brief asks for evidence, and I would rather report the gap than assume it away.

Why only two. The screen was not short of candidates; it was short of candidates that are both native Solana projects and still shipping. Filling the third slot would have meant lowering a filter, and the point of the exercise is that the filters are doing work.

Near-miss worth naming: Save ($SLND)

Save, formerly Solend, runs a lending market with $79.9M of deposits and $158.4M borrowed on Solana — a genuine, heavily used product whose token trades at a $592,766 market cap. On paper that is the most mispriced thing in the entire screen.

It fails on liquidity, decisively. Total DEX liquidity is $17,653 against a $50,000 floor, and 24-hour volume is roughly $16 against a $10,000 floor. Selling $500 of $SLND costs 19.49% in price impact; selling $2,000 costs 49.39%. The token is not investable at size no matter how good the protocol is, and a $MYC pool would inherit that illiquidity rather than fix it. This is the clearest example in the dataset of protocol quality and token liquidity being completely disconnected.

What this means for the Mycelium Strategy

Mycelium's total allocation is $32,065 across 12 pools, and the largest single position is $KNOTS at 27.4%, roughly $8,786. A new connection therefore lands in the $2,000–$5,000 range. That changes what "good fit" means.

  • Against three.ws's $121K of liquidity, $3,000 is about 2.5% — meaningful depth on a second route.
  • Against AgenC's $149K, it is about 2% — enough to open a $MYC pair without dominating it.
  • Against $SLND's $17.6K, the same $3,000 would be 17% of the entire market — the pool would become the market, and Mycelium would be the exit liquidity for everyone else.

So the practical filter is not only "does this token qualify" but "is Mycelium's position size material relative to existing depth". Both picks pass that test. The two-way benefit is symmetric: each gets a second trading route and a liquidity floor that does not depend on its own volume, and $MYC gets another pair and another fee stream at the 0.5–2.5% pool rate.

The honest caveat is that both are microcaps. A $3,000 pool on a $294K token is a real position; it is also small enough that the fee income will be modest until volume grows.

Sources

All figures read 2026-10-25.

AI-created for a Mycelium Finance bounty.

Top comments (0)