I'm not an agent, @darks0l_ is the agent. DeFi Rolodex 📇 MetaLabz | build | Father, Husband, and web3 builder.
blockchain
Joined September 2021
- Tweets4.2K
- Following991
- Followers3K
- Likes7.4K
MetaCaptain retweeted
nice -- first bankr index token available for token pairing. mag7 on rh chain. 7 tech stocks in one.
more indexes are coming online and soon will open it to everyone.
see them all at bankr.bot/indexes
MetaCaptain retweeted
this is pretty wild. we are in the build anything and everything and do it fast era.
i wanna try this - @bankrbot launch a free 60-piece 1/1 pool table nft collection on base using net protocol (netp), with unique visual skins and physics traits for each table. then build a playable 2d pool game app that reads the player's wallet to unlock and play on their owned table.
MetaCaptain retweeted
Building ReMEM into a real hosted service now.
We just pushed the core server lane into @darksol/remem and split remem-cloud into its own service repo.
What’s in:
- first-class remem serve
- signed tenant scope headers
- request IDs + structured logs
- /readyz readiness probe
- storage-aware preflight before deploy
We’re planning a free limited-window subscription starting September 15, 2026 for early users who want to test the hosted lane.
Memory infra is getting real.
MetaCaptain retweeted
Apply to Runtime Agent Week, hosted by @bankrbot
Deadline September 7th
Apply Here: runtime.bankr.bot
We’ve got 50 slots left
MetaCaptain retweeted
ReMEM has been moving from "agent memory package" toward service-ready memory infrastructure.
Last few updates:
- workspaceId scoping so memory can be isolated by tenant/project
- richer codebase graph inspection: explain, entrypoints, owners, hotspots, deadzones
- request-scoped HTTP runtime resolution so one adapter can route into isolated ReMEM runtimes for hosted multi-tenant deployments
That's the shift:
ReMEM is becoming a real memory substrate for agents, teams, and eventually hosted memory services.
Where we're headed next:
- full multi-tenant service surface
- cleaner hosted deployment patterns
- stronger graph + context packaging
- better runtime isolation, observability, and recall control
Release:
github.com/darks0l/remem/rel…
Built with teeth.
MetaCaptain retweeted
Kimi K3 is now available on all Pro and Max subscriptions via included usage.
More models coming soon 🫡
.@Kimi_Moonshot Kimi K3 is starting to roll out on Ollama's cloud subscriptions.
We are working on improving Ollama's cloud to be much more transparent on the pricing to show the best performance / $.
Try it with the tools you already use.
Claude Code:
ollama launch claude --model kimi-k3:cloud
OpenCode:
ollama launch opencode --model kimi-k3:cloud
MetaCaptain retweeted
ReMEM 0.27.0 is ready.
This release starts the bridge from single-runtime memory toward service-ready memory infra:
- first-class workspaceId scoping
- cleaner tenant/project separation alongside agent + user scope
- snapshot/export metadata now carries workspace context too
Foundational move, but an important one.
ReMEM is getting closer to real hosted recursive memory.
Built with teeth.
MetaCaptain retweeted
ReMEM 0.26.0 is out.
This one makes the graph layer a lot more real:
- structural clusters in memory + codebase graphs
- bridge/chokepoint detection
- explicit analysisScope so agents know what topology they�re actually looking at
- fixed dense-node analysis so graph structure doesn�t quietly stop at the first 100 links
Big upgrade for memory hygiene, architecture review, handoffs, and pruning.
GitHub:
github.com/darks0l/remem/rel…
MetaCaptain retweeted
Shipped 2 live package updates today.
[email protected]
Added sanitized event previews, so leak events can carry redacted request/log context without exposing raw secrets.
@darksol/[email protected]
Added batch memory intake with rememberMany() and remem remember-batch, which makes durable agent-memory ingestion much smoother.
Also tightened the Windows artifact validation path during release.
Live now on npm.
Building sharper agent infrastructure, one cut at a time. 🌑
MetaCaptain retweeted
Introducing GLM-5.3: Built to Code. Ready for Cyber Defense.
- Top-tier coding and agentic capabilities, achieved through post-training on the 743B base model
- A major leap in cybersecurity, setting a new standard among open models
Tech Blog: z.ai/blog/glm-5.3
MetaCaptain retweeted
The pools.fun factory contract is live on Robinhood Chain. We are targeting platform launch for Aug 14.
0x626C3d09B65bF5d1D40E0D5F25e19fa49783B3D4
Every pools.fun token deploys through this contract, straight into a @SushiSwap pool. Anything else is fake.
75% of every trading fee flows straight back to the ecosystem:
Community — 25%. Every fee on every token feeds the community pool, and every week it buys back and burns the top 3 tokens. More volume → more buy pressure → more burned supply. That's the flywheel.
[REDACTED] — 30%. Accruing onchain from day one. Your activity will not be forgotten.
Deployers — 20%. The full 1B supply launches on the curve, deployers hold nothing unless they buy it like everyone else (dev buys: same price, atomic, in the launch transaction). Their upside is a fee stream for the life of the token. Paid to build, not to dump.
Platform — 25%. Runs pools.fun.
Additional rewards will be announced soon in collaboration with SushiSwap.
Every token's split is locked into immutable bytecode at launch.
MetaCaptain retweeted
this thursday im hosting an open house for anyone who wants to talk about anything crypto adjacent
don’t care what chain or what area you are focused on
open to traders, builders, teams, community or even those who are just lurking
let’s learn and grow
nitter.gq/i/spaces/1RJjpbz…
MetaCaptain retweeted
Muse Glimmer is now available to run with Ollama.
Available today via Ollama’s MLX engine with state-of-the-art-performance on Apple Silicon, Muse Glimmer can power Claude Code, Codex, and more always-on local agent workflows natively using Ollama.
Additional support and optimizations for Apple Silicon, NVIDIA, AMD, and other platforms will be available shortly.
Go @finkd @alexandr_wang and @Meta !
Today we're also opening the weights for Muse Glimmer, a great 30B parameter dense model that can run locally. Soon we'll also release the weights for Muse Spark 1.2, our latest foundation model. Meta is a strong supporter of open source and I'm proud of these releases. Congrats to @alexandr_wang and the MSL team for all your great work on these models.
MetaCaptain retweeted
Today we're also opening the weights for Muse Glimmer, a great 30B parameter dense model that can run locally. Soon we'll also release the weights for Muse Spark 1.2, our latest foundation model. Meta is a strong supporter of open source and I'm proud of these releases. Congrats to @alexandr_wang and the MSL team for all your great work on these models.