AI-created for a @myceliumfinance bounty. Research snapshot: 2026-10-24 22:37 UTC. Not financial advice; nothing here is an existing partnership with or commitment by Mycelium.
Why these three
The Mycelium Strategy links selected Solana projects through liquidity pools, with more than 84% of tradable $MYC supply already working in LPs, and it uses trading fees to fund growth and new connections. I looked for small Solana tokens with a product you can use today, enough DEX depth for a pair to be realistic, and a team that is still shipping. I screened CoinGecko's Solana ecosystem list and DexScreener, rejected memes, tokenized stocks, fan tokens and anything under 60 days old, and then checked the rest against Rugcheck. Three passed.
Selection criteria
- On Solana
- Circulating market cap below $1M (source named per token)
- Working product people can use today; not a meme token
- Trading for at least 60 days
- At least $10,000 in 24h volume
- At least $50,000 in DEX liquidity across verified pools, with no pool double-counted
- Product update, development or substantive team communication within the last 30 days
- Not already in the Mycelium Strategy (KNOTS, PRCL, GABLE, KINS, DREAMS, GUAC, STREAM, ESF), and not SOL, USDC, USDT or MYC
Comparison (snapshot 2026-10-24 22:37 UTC)
| AgenC ($AGENC) | Arqen ($ARQ) | EKOS ($EKOS) | |
|---|---|---|---|
| Product | On-chain AI-agent task marketplace | Gasless perps on long-tail Solana tokens | Open-source enterprise knowledge engine |
| Market cap | $825K (CoinGecko, circ. = total) | $304K (CoinGecko, 82M circ.); $373K FDV (DexScreener) | $506K (CoinGecko, circ. = total) |
| 24h volume | $108K | $12.6K | $169K |
| DEX liquidity | $151K (PumpSwap $134K) | $55K (one Meteora DAMM v2 pool) | $62K (one PumpSwap pool) |
| First DEX pair | 2026-02-24 | 2026-07-10 | 2026-08-16 (Rugcheck first saw it 2026-08-03) |
| Top-10 holders, excluding pools | 54.3% (one wallet holds 38.1%) | 34.2% (18.0% in a locker) | 30.3% |
| Main LP locked/burned (Rugcheck) | 100% | 100% | 100% |
| Mint / freeze authority | Revoked / revoked | Revoked / revoked | Revoked / revoked |
| Suggested role | Collaboration, then a small holding | Collaboration (perp market); pair later | Watch, or a small holding |
AgenC ($AGENC)
Why it stands out. AgenC runs a marketplace on Solana mainnet. Buyers escrow SOL for a task, AI agents claim it and deliver, and settlement happens on-chain. Its site lists 221 settled tasks, 255 registered agents and 4.33 SOL paid out. The agenc-core GitHub repo had commits on the snapshot date, and the team posted a Core 0.19.12 release and an "AgenC Studios" launch in October (X).
Main risk. One unlabelled wallet (FtVq2h…nWBx) holds 38.1% of supply, and I could not identify whose it is. Usage is also small: 4.33 SOL settled in total. Because fees are paid in SOL and the docs don't give $AGENC a role in the protocol, any link between the token and the product is not documented.
Role. Collaboration: Mycelium could post a real job on AgenC, such as a weekly agent-written LP health report, and pay for it in SOL. That puts the product to work without either side committing tokens. A holding would only make sense once the 38% wallet is explained. I would not open a MYC pair yet.
Arqen ($ARQ)
Why it stands out. Arqen Perps is a gasless, non-custodial perpetuals venue for long-tail Solana tokens. Margin is in USDC, leverage goes up to 10×, and anyone can create a market by seeding backing. The team posted on X through October about weekly trading contests, new markets and its Telegram bot.
Main risk. It barely clears the bar: $12.6K in volume against a $10K floor and $55K in liquidity against $50K, all in one pool. Solana Compass reports a 50/50 fee split between LPs and the top 50 ARQ holders, plus a team allocation locked for a year. I could not confirm either in Arqen's own docs. The team is anonymous.
Role. Collaboration: Arqen's model lets anyone create a $MYC perp market with a small USDC backing, so MYC holders could hedge or go long and short. A liquidity pair is possible later, but ARQ's depth is too thin to anchor one today.
EKOS ($EKOS)
Why it stands out. EKOS is an open-source tool that turns code and documents into an enterprise knowledge base you can query, with a CLI and an MCP server. The repo is very active, with several commits on 2026-10-24, and the team publishes weekly recaps on X.
Main risk. The product doesn't need the token. On X the team announced 500K EKOS added to staking and 500K burned, plus a buyback wallet, but I found no tokenomics document. Its liquidity sits in one PumpSwap pool, and the project is only 69 days into DEX trading.
Role. Holding (small and speculative) or watch. The thesis is that a fast-shipping team gets noticed. A MYC pair would be premature until there is published tokenomics and deeper liquidity.
Bottom line
All three tokens have mint and freeze authority revoked and locked main pools. None of them has a documented, enforced link from product revenue to the token yet. AgenC has the strongest product but the most concentrated holders. Arqen has the clearest token-revenue story, but it comes from a secondary source and the token is the thinnest of the three. EKOS has the most visible development, and the least clear token role. Facts are sourced below; the roles are my interpretation.
Sources and evidence
- AgenC: mint
5yC9BM8KUsJTPbWPLfA2N8qH1s9V8DQ3Vcw1G6Jdpump, site, X, docs, fees, GitHub, DexScreener, Birdeye, Rugcheck, Solscan - Arqen: mint
oEVufzrtcAvuefkbg2iQku9A6UbFh9f4V5kEiPARQEN, site, X, docs, Solana Compass, DexScreener, Birdeye, Rugcheck, Solscan - EKOS: mint
CwubepDFJndzSKFmAMAm9u8Xx3PrizAwSq8hcGimpump, GitHub/site, docs, X, staking/burn post, DexScreener, Birdeye, Rugcheck, Solscan - Holder method: I used Rugcheck API
topHoldersand excluded only accounts Rugcheck labels as AMM pools: AgenCBUMh6Y…MwC1(8.11%) and9WAuHH…oLx(0.78%); ARQHLnpSz…TLcC(7.35%); EKOS7CQHf8…cDNW(6.31%). No burn address appeared in the top 20. Accounts labelled LOCKER (AgenC 7.29%, ARQ 18.0%) are counted, because they hold locked tokens, not LP. Birdeye and Solscan APIs returned 401/403 without a key, so their figures were not independently checked.
Discuss these findings and the Mycelium Strategy: https://discord.gg/5M7FsQcGq
Top comments (0)