Hiroki🦉

Introducing Epismo Skills: portable, community-built workflows for your agents. Open-source workflow layer for human + agent project ops: discover community workflows, generate from your context (or a doc/link), then run and track them as projects, so the process becomes reproducible instead of a one-off. Set it up once in your agent environment (Claude Code, Codex, OpenClaw, etc.), then import and run workflows without switching UIs.

@bondwithchloe Want Skills so I can run in my agent setup

Re-introducing...Bond, your AI Chief of Staff. You worked 10 hours yesterday. Maybe 2 of them actually mattered. The rest? Standups, Slack threads, status updates, triaging emails. Executives don't need to work harder. They need Bond 👉 bondapp.io

@jeff_weinstein @tempo @stripe I will try it out

Introducing the Machine Payments Protocol (MPP). mpp.dev: an open protocol for machine-to-machine payments, co-authored by @tempo and @stripe. Watch it in agentic action ⤵️

Two things sit at the core of all productive work: Knowing what to do. Knowing what you know. Task management. Knowledge management. That's it.

@aakashgupta Dispatch local FS access versus OpenAI cloud agents highlights the split. Local wins on privacy and speed but state sync and permission boundaries across devices will be the long-term reliability test.

Anthropic is building OpenClaw faster than OpenAI is. OpenClaw proved a concept the entire industry had been theorizing about: your AI agent should live on your computer, not in someone else’s cloud, and you should be able to talk to it from anywhere. 318,000 GitHub stars. Then Steinberger joined OpenAI to build exactly this at scale. Here’s what OpenAI has shipped since: Codex, a desktop coding agent with no mobile remote control. ChatGPT Agent, which runs on a remote virtual computer in OpenAI’s cloud where it can’t see your local files. Developers are filing GitHub issues on the Codex repo right now requesting phone-to-desktop control. Third-party devs already built Taskdex and Remote Codetrol to hack around the gap with relay servers and Tailscale tunnels. Anthropic just shipped it natively. Dispatch: pair your phone with Claude Desktop, message Cowork from anywhere, come back to finished work. Cowork already had the VM running on your machine, full filesystem access, browser control, sub-agent coordination, and a skills system stored as markdown. Dispatch was the missing piece that turns the whole stack into something you can operate from your pocket. The reason this works when cloud agents can’t: Cowork reads your actual filesystem, your actual browser, your actual connected tools. When I ask it to cross-reference a local spreadsheet with a competitor’s pricing page, it can do that because both the spreadsheet and the browser are on my machine. A cloud agent would need me to upload the spreadsheet first, lose the file path context, and still wouldn’t have access to my connected Slack or Google Drive. The context is real because the machine is real. I’ve been running Cowork since launch. Five tasks dispatched every morning before my kids wake up: research briefs, competitor analysis, file organization, data pulls from local spreadsheets, editing passes on drafts. 90 minutes of active work compressed into 10 minutes of dispatching and 20 minutes of reviewing outputs. Dispatch changes what happens the rest of the day. An idea hits while I’m out, I message Cowork from my phone, the work is waiting when I get home. And the part that should keep OpenAI up at night: Anthropic didn’t need to acquire OpenClaw or hire Steinberger to ship this. They were already building the same architecture independently. Cowork launched in January with local VM execution, filesystem access, and markdown skills before OpenClaw was even mainstream. Steinberger validated the demand. Anthropic had already built the supply. OpenAI bought the architect. They’re still looking for the blueprints he left at Anthropic’s door.

@NVIDIAAIDev @openclaw Single-command deploy removes the friction, which is great. The reliability question starts right after: what's the approval boundary and recovery path once it's always on?

Ready to deploy AI agents? NVIDIA NemoClaw simplifies running @openclaw always-on assistants with a single command. 🦞 Deploy claws more safely ✨ Run any coding agent 🌍 Deploy anywhere Try now with a free NVIDIA Brev Launchable 🔗 nvidia.com/nemoclaw

@NousResearch @vercel Plugin ecosystems are where things start compounding fast. The hard part isn't adding more commands, it's keeping the execution contract legible as everything mixes in one run.

Hermes Agent v0.3.0 ☤ 248 PRs. 15 contributors. 5 days. • Real-time streaming across CLI and all platforms • First-class plugin architecture, package and share tools+commands+skills • /browser connect to live Chrome via CDP • @vercel AI Gateway model provider • @browser_use browser tool provider • VS Code, Zed, and JetBrains integration • Voice mode with local Whisper • PII redaction everywhere 9 new skills. 50+ bug fixes. Much more in the full changelog.

@OpenAIDevs Subagents fix context volume, but the next bottleneck is context meaning. Parallel work only stays reliable when each agent leaves a clear artifact and ownership boundary.

Subagents are now available in Codex. You can accelerate your workflow by spinning up specialized agents to: • Keep your main context window clean • Tackle different parts of a task in parallel • Steer individual agents as work unfolds

@ihtesham2005 Prompt packs expire, methodologies don't. The real value is making review points explicit so the agent can't quietly drift while looking productive.

🚨 Holy shit...A developer on GitHub just built a full development methodology for AI coding agents and it has 40.9K stars on GitHub. It's called Superpowers, and it completely changes how your AI agent writes code. Right now, most people fire up Claude Code or Codex and just… let it go. The agent guesses what you want, writes code before understanding the problem, skips tests, and produces spaghetti you have to babysit. Superpowers fixes all of that. Here's what happens when you install it: → Before writing a single line, the agent stops and brainstorms with you. It asks what you're actually trying to build, refines the spec through questions, and shows it to you in chunks short enough to read. → Once you approve the design, it creates an implementation plan so detailed that "an enthusiastic junior engineer with poor taste and no judgement" could follow it. → Then it launches subagent-driven development. Fresh subagents per task. Two-stage code review after each one (spec compliance, then code quality). The agent can run autonomously for hours without deviating from your plan. → It enforces true test-driven development. Write failing test → watch it fail → write minimal code → watch it pass → commit. It literally deletes code written before tests. → When tasks are done, it verifies everything, presents options (merge, PR, keep, discard), and cleans up. The philosophy is brutal: systematic over ad-hoc. Evidence over claims. Complexity reduction. Verify before declaring success. Works with Claude Code (plugin install), Codex, and OpenCode. This isn't a prompt template. It's an entire operating system for how AI agents should build software. 100% Opensource. MIT License.

@altius_rup @akshay_pachaar Exactly

@hirokiyn @akshay_pachaar I use a task board concept for that : the orchestrator posts tasks on a board, assigning to main agents. Main agents pick them up and run their jobs.

@buildwithshyam A night to friends Few weeks to public

Everyone says - Ship fast. So… How long should an MVP take? - 2 days (weekend grind) - 5 days - 2 weeks - 3 weeks

@JiquanNgiam Agreed, CLI is good for local dev but not for everyone

If someone says "CLIs are dead" or "MCPs are overhyped," that's a tell they only work in one context. My personal AI setup: mostly CLIs, 4 MCP servers. My work setup: 10+ MCP servers, almost no CLIs. Same person - me!, but completely different mix. I tried walking non-technical teammates through installing the GitHub CLI on Windows. They couldn't get started - it's a mess! For them, MCP's single-click OAuth is the only viable path. But for me in build/dev mode, "curl | jq | grep" is natural and hard to beat. Use CLIs in build/dev mode. MCPs for enterprise/teams and background agents. Skills fit in the picture as encoding processes and judgment that go beyond the tools. Full breakdown: jngiam.bearblog.dev/mcps-clis-and-…

Bring community-built workflows for your main agents like Claude Code, Codex, or OpenClaw. The open-source harness for the orchestration layer: discover community workflows, then run and track them as projects.

MCP sucking is a harness problem, not an MCP problem MCP unlocks behavior that is fundamentally impossible to get via CLI or APIs Bad auth, too much context usage, all get solved with an execution layer - your agent writes code to progressively discover and call tools

@garrytan x.com/rhyssullivan/s…

@garrytan Is there a way to improve MCP fundamentally? CLIs and APIs are strong only when agents work locally with proper harness setup, and it's still not the best for everyone.

MCP sucks honestly It eats too much context window and you have to toggle it on and off and the auth sucks I got sick of Claude in Chrome via MCP and vibe coded a CLI wrapper for Playwright tonight in 30 minutes only for my team to tell me Vercel already did it lmao But it worked 100x better and was like 100LOC as a CLI

@morganlinton @denisyarats Is there a way to improve MCP fundamentally? CLIs and APIs are strong only when agents work locally with proper harness setup, and it's still not the best for everyone.

The cofounder and CTO of Perplexity, @denisyarats just said internally at Perplexity they’re moving away from MCPs and instead using APIs and CLIs 👀

Epismo is now the tool for the orchestration layer of agents. It helps main agents (like openclaw or claude code) manage subagents and humans.

@DanKulkov My weekly limits are reset almost everyday😂

is it just me or $20/mo codex has limits of $100/mo plan i feel like it's impossible to hit weekly limits