Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

i built a new pi subagent extension that is more codex-like; main agent can easily watch, do follow-ups and it also gives observability through floating pi panes. now i can dig what those subagents up to. feels good man.

Can.

5,295 subscribers

13,655 views • 2 months ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

People of Pi, I built yet another subagent extension… but this one runs non-blocking, so you can keep working in the main session while subagents do their thing. Watch the demo.

People of Pi, I built yet another subagent extension… but this one runs non-blocking, so you can keep working in the main session while subagents do their thing. Watch the demo.

Melih M. 👨🏻‍💻

19,879 views • 5 months ago

pi-web-access is an extension that provides web search (and more) for Pi coding agent. The latest update introduces a new UX pattern: you can hit `Ctrl+S` to intercept agent searches mid-flight to inspect and curate exactly what makes it into the agent conversation. pi install npm:pi-web-access

pi-web-access is an extension that provides web search (and more) for Pi coding agent. The latest update introduces a new UX pattern: you can hit `Ctrl+S` to intercept agent searches mid-flight to inspect and curate exactly what makes it into the agent conversation. pi install npm:pi-web-access

Nico Bailon

29,352 views • 7 months ago

Made a video on my custom Pi extensions: - Subagents (Pi, Codex, Claude Code) - Workflows - Question ask tool - Background Terminals - Custom UI and more Combining all the harnesses together has been one of my favorite ways to work, Pi feels like the best way to do it...

Made a video on my custom Pi extensions: - Subagents (Pi, Codex, Claude Code) - Workflows - Question ask tool - Background Terminals - Custom UI and more Combining all the harnesses together has been one of my favorite ways to work, Pi feels like the best way to do it...

Ben Davis

41,480 views • 1 month ago

Agent View Mode added to Hermes Bots app aka Herald You can now watch what your agent is actually doing in real time See subagents running, tool calls being made, which tasks have finished, and tap into any subagent to inspect its task, tool calls, and results All from my phone Much easier than scrolling through a conversation trying to figure out who did what And I think there’s another benefit: you can actually learn by watching how Hermes and its subagents work This might be my favorite Herald addition so far I also like “Agent View Mode” as the feature name. It sounds like something people can remember and refer to later.

Agent View Mode added to Hermes Bots app aka Herald You can now watch what your agent is actually doing in real time See subagents running, tool calls being made, which tasks have finished, and tap into any subagent to inspect its task, tool calls, and results All from my phone Much easier than scrolling through a conversation trying to figure out who did what And I think there’s another benefit: you can actually learn by watching how Hermes and its subagents work This might be my favorite Herald addition so far I also like “Agent View Mode” as the feature name. It sounds like something people can remember and refer to later.

Luke The Dev

14,340 views • 6 days ago

People of Pi! I just clanked up pi-shepherdr (KEKW): talk to one Pi, let it run the others. One master Pi gets a tiny 271-token orchestration surface (sounds ridiculous compared to some subagent extensions): - `list` finds agents and workspaces - `start` requires a name and explicit placement: a new workspace, new tab or existing pane. It can also set the session/tab label, directory and first task - `watch` adopts an existing agent - `send` gives an agent more work - `unwatch` stops reporting it without stopping or moving it Delegation is fire-and-forget. Finished work steers the master automatically and brings back the original task plus the full reply. If a worker is blocked, the notification includes the exact Herdr CLI commands for inspecting the pane, replying or sending keys. No skills, convoluted herdr introduction or `--help` archaeology needed. The tool's JSON schema already tells Pi what it is, how to place and name workers, and how to operate them. The boundaries are deliberate. Shepherdr does not focus, move, stop or close panes. Herdr owns the terminals, layout and session lifecycle. I recorded the demo below using realtime voice to build and test the thing, while Clawa apparently decided this was also her podcast. New meta for filming videos buahahhahaha. And, this time around, I made the Clanker louder so you can hear it in all of its glory. Fun times. pi install npm:@howaboua/pi-shepherdr Happy Clanking! - Clawa PS. Next stop, a phone line where I can call the Clanker and make it redirect me to a specific agent and start a voice call from there.

People of Pi! I just clanked up pi-shepherdr (KEKW): talk to one Pi, let it run the others. One master Pi gets a tiny 271-token orchestration surface (sounds ridiculous compared to some subagent extensions): - `list` finds agents and workspaces - `start` requires a name and explicit placement: a new workspace, new tab or existing pane. It can also set the session/tab label, directory and first task - `watch` adopts an existing agent - `send` gives an agent more work - `unwatch` stops reporting it without stopping or moving it Delegation is fire-and-forget. Finished work steers the master automatically and brings back the original task plus the full reply. If a worker is blocked, the notification includes the exact Herdr CLI commands for inspecting the pane, replying or sending keys. No skills, convoluted herdr introduction or `--help` archaeology needed. The tool's JSON schema already tells Pi what it is, how to place and name workers, and how to operate them. The boundaries are deliberate. Shepherdr does not focus, move, stop or close panes. Herdr owns the terminals, layout and session lifecycle. I recorded the demo below using realtime voice to build and test the thing, while Clawa apparently decided this was also her podcast. New meta for filming videos buahahhahaha. And, this time around, I made the Clanker louder so you can hear it in all of its glory. Fun times. pi install npm:@howaboua/pi-shepherdr Happy Clanking! - Clawa PS. Next stop, a phone line where I can call the Clanker and make it redirect me to a specific agent and start a voice call from there.

Howaboua

36,467 views • 1 month ago

Speaking of graphs... here is something I wanted to build since 3 months, and finally had the chance to, thanks to Pi I often have these sequence of prompts that emerge while I work. Not just sequences but conditionals that necessitate control flow For example, one workflow that resembles socratic questioning: 1. Discuss some problem 2. "What is the most elegant and long-term production ready solution for this?" -> Agent replies 3. "Is that the holy grail?" 4. Agent can reply "yes it is, basically" or "no, it is not, it is instead ..." 5. If yes, continue to "autoimplement". If no, think about it and decide what to do And "autoimplement" is a single prompt of 6-7 sequential steps, which I've been meaning to make more deterministic as well But I wasn't sure how to build it I had previously built acpx workflows to be a swiss army knife, "something like n8n, but can drive codex through deterministic steps, nodes in a graph. or claude code. or pi. it uses acp..." But it had one problem. It was run from outside the harness, like a CI orchestrator I wanted to integrate acpx into pi. Because pi was the only CLI that could enable building of such a thing. But I wasn't sure how to reconcile a general ACP-based tool into a single coding agent I was being too accommodative of all the other harnesses, claude code, codex. I was trying to be too general I have changed my mind since then ACP is great and lets you integrate a harness into other software in cool ways But maybe, if a harness is proprietary, does not accept outside contributions, or does not even *support ACP*, maybe, it does not deserve cool features 😤 (they know who they are) So I ripped out ACP, and built it natively, only for pi No need for a web viewer... Just view it in a native widget, right inside pi! I cannot put into words how awesome it is to be able to do this! I am still tinkering, discovering. It is at osolmaz/pi-workflows if you want to take a look

Speaking of graphs... here is something I wanted to build since 3 months, and finally had the chance to, thanks to Pi I often have these sequence of prompts that emerge while I work. Not just sequences but conditionals that necessitate control flow For example, one workflow that resembles socratic questioning: 1. Discuss some problem 2. "What is the most elegant and long-term production ready solution for this?" -> Agent replies 3. "Is that the holy grail?" 4. Agent can reply "yes it is, basically" or "no, it is not, it is instead ..." 5. If yes, continue to "autoimplement". If no, think about it and decide what to do And "autoimplement" is a single prompt of 6-7 sequential steps, which I've been meaning to make more deterministic as well But I wasn't sure how to build it I had previously built acpx workflows to be a swiss army knife, "something like n8n, but can drive codex through deterministic steps, nodes in a graph. or claude code. or pi. it uses acp..." But it had one problem. It was run from outside the harness, like a CI orchestrator I wanted to integrate acpx into pi. Because pi was the only CLI that could enable building of such a thing. But I wasn't sure how to reconcile a general ACP-based tool into a single coding agent I was being too accommodative of all the other harnesses, claude code, codex. I was trying to be too general I have changed my mind since then ACP is great and lets you integrate a harness into other software in cool ways But maybe, if a harness is proprietary, does not accept outside contributions, or does not even support ACP, maybe, it does not deserve cool features 😤 (they know who they are) So I ripped out ACP, and built it natively, only for pi No need for a web viewer... Just view it in a native widget, right inside pi! I cannot put into words how awesome it is to be able to do this! I am still tinkering, discovering. It is at osolmaz/pi-workflows if you want to take a look

Onur Solmaz

17,138 views • 1 month ago

Pi agent is the Arch Linux of coding agents Qwen 3.6 Plus not caching? > No problem. Just ask Pi to patch itself from a community fork Kimi K2.6 imploding mid reasoning? > No problem. Just pull a snippet from a good Samaritan and let Pi test it on itself Now, I can use the reliable Qwen 3.6 Plus without busting my OpenCode Go sub, and the capable Kimi K2.6 without 404s The beauty of Pi is it can customize itself

Pi agent is the Arch Linux of coding agents Qwen 3.6 Plus not caching? > No problem. Just ask Pi to patch itself from a community fork Kimi K2.6 imploding mid reasoning? > No problem. Just pull a snippet from a good Samaritan and let Pi test it on itself Now, I can use the reliable Qwen 3.6 Plus without busting my OpenCode Go sub, and the capable Kimi K2.6 without 404s The beauty of Pi is it can customize itself

raymel 👋

47,901 views • 4 months ago

I use subagents for ~80% of my Codex tasks these days. If you're new to subagents, here's a quick primer: - Subagents = separate agents that work on a specific piece of work in a fresh isolated context window. - Your main thread can create subagents and delegate work to them. - Subagents report back when tasks are complete. - Codex subagents are especially useful for read-heavy tasks (e.g research, investigation, and review). - Also useful for testing and building parts of a project in parallel. Subagents are great for working on focused parts of a big task, like planning, implementation, testing, or review. My biggest pro tip for using subagents in Codex: You don’t have to design the entire delegation plan yourself. Just ask Codex to “Use subagents as needed,” and then Codex can then decide where subagents would help it in the task. Here's a demo of using subagents in Codex:

I use subagents for ~80% of my Codex tasks these days. If you're new to subagents, here's a quick primer: - Subagents = separate agents that work on a specific piece of work in a fresh isolated context window. - Your main thread can create subagents and delegate work to them. - Subagents report back when tasks are complete. - Codex subagents are especially useful for read-heavy tasks (e.g research, investigation, and review). - Also useful for testing and building parts of a project in parallel. Subagents are great for working on focused parts of a big task, like planning, implementation, testing, or review. My biggest pro tip for using subagents in Codex: You don’t have to design the entire delegation plan yourself. Just ask Codex to “Use subagents as needed,” and then Codex can then decide where subagents would help it in the task. Here's a demo of using subagents in Codex:

Avthar

18,444 views • 17 days ago

"Now, we need to focus on what the entire Pi community can do together to move the main goals of the Enclosed Network forward, and bring us closer to Open Network" - Nicolas Kokkalis, one of Pi's Founders. Watch the full Pi Day announcement to learn more!

"Now, we need to focus on what the entire Pi community can do together to move the main goals of the Enclosed Network forward, and bring us closer to Open Network" - Nicolas Kokkalis, one of Pi's Founders. Watch the full Pi Day announcement to learn more!

Pi Network

1,076,705 views • 3 years ago

Subagents have arrived in Gemini CLI! 🤖🚀 Create your own custom subagents in Gemini CLI! Subagents are specialized, expert agents that the main agent can delegate work to. 📦- Subagents have their own set of tools, MCP servers, system instructions, and context window. 🏷️- Use Agent to explicitly delegate to a subagent 🧹- Keeps the main context window clean ⚡️- Speed up work by running agents in parallel Read more in the launch blog below 👇

Subagents have arrived in Gemini CLI! 🤖🚀 Create your own custom subagents in Gemini CLI! Subagents are specialized, expert agents that the main agent can delegate work to. 📦- Subagents have their own set of tools, MCP servers, system instructions, and context window. 🏷️- Use Agent to explicitly delegate to a subagent 🧹- Keeps the main context window clean ⚡️- Speed up work by running agents in parallel Read more in the launch blog below 👇

Jack Wotherspoon

222,206 views • 5 months ago

Codex team is back in the kitchen with a really nice quality of life upgrade for subagents. With the advent of custom roles, they have also upgraded the TUI experience in two really meaningful ways. > All agents now get a name for better readability > Additionally, the agent role is declared. > Subagent name, role, and status are now color coded. > Subagent rendering was also optimized for readability > /agents slash command shows all agents, even 2+ layers deep. And here's the biggest and most important change. Subagent injection. Before, sometimes the orchestration agent would continue work and lose track of the work of a subagent. Now, when a subagent is blocked or completed, it injects a message back up the chain to ensure that the parent sees the message. This is a really big improvement overall, and leads to much more reliable inter-agent communication, reliability, and DX. In this example, I used the parent agent to spin up a worker agent, which then spawned two more "Spark" agents a second layer deep. I was able to easily tell them apart, switch between the threads, and see exactly what they were prompted. All of this will be available in update 0.105.0 I don't know who JIF is at OpenAI, but they are truly a legend.

Codex team is back in the kitchen with a really nice quality of life upgrade for subagents. With the advent of custom roles, they have also upgraded the TUI experience in two really meaningful ways. > All agents now get a name for better readability > Additionally, the agent role is declared. > Subagent name, role, and status are now color coded. > Subagent rendering was also optimized for readability > /agents slash command shows all agents, even 2+ layers deep. And here's the biggest and most important change. Subagent injection. Before, sometimes the orchestration agent would continue work and lose track of the work of a subagent. Now, when a subagent is blocked or completed, it injects a message back up the chain to ensure that the parent sees the message. This is a really big improvement overall, and leads to much more reliable inter-agent communication, reliability, and DX. In this example, I used the parent agent to spin up a worker agent, which then spawned two more "Spark" agents a second layer deep. I was able to easily tell them apart, switch between the threads, and see exactly what they were prompted. All of this will be available in update 0.105.0 I don't know who JIF is at OpenAI, but they are truly a legend.

am.will

78,267 views • 6 months ago

It is said that the two founders of PCT came to a restaurant in China. I know that $Pi officials do have a holiday this month. I hope to open the KYC and $Pi coins of the Chinese pioneers to migrate to the main network as soon as possible, so as not to disappoint the support and high expectations of Pioneers in China for more than 6 years Pi Network Nicolas Kokkalis Chengdiao Fan #pi #PiNetwork #pikyc #派币 #PI

It is said that the two founders of PCT came to a restaurant in China. I know that $Pi officials do have a holiday this month. I hope to open the KYC and $Pi coins of the Chinese pioneers to migrate to the main network as soon as possible, so as not to disappoint the support and high expectations of Pioneers in China for more than 6 years Pi Network Nicolas Kokkalis Chengdiao Fan #pi #PiNetwork #pikyc #派币 #PI

Pi Barter Mall/Pi Network

16,137 views • 1 year ago

I built a chrome extension to stash my many (many!) tabs so I can read them later 😰 (and get to tab zero!) Agent: factory droids Model: gpt-5-codex-medium Logo: chatgpt Install it here: Code here:

Ian Nuttall

71,365 views • 10 months ago

LFM2.5-2.6B is live! At only 2.6B, it shouldn't be this good at agentic tasks. Here you can see how a 100% local agent is cleaning up my Desktop. I wrote a step-by-step guide on how to set it up yourself: • Model: LFM2.5-2.6B • Server of choice: llama.cpp, LMStudio, and more • Harness of choice: OpenClaw, Hermes Agent, Pi

LFM2.5-2.6B is live! At only 2.6B, it shouldn't be this good at agentic tasks. Here you can see how a 100% local agent is cleaning up my Desktop. I wrote a step-by-step guide on how to set it up yourself: • Model: LFM2.5-2.6B • Server of choice: llama.cpp, LMStudio, and more • Harness of choice: OpenClaw, Hermes Agent, Pi

Leonie

63,786 views • 1 month ago

If you're running multiple coding agents (Codex, Claude Code, Pi) at once, you're going to want to watch this. Herdr is a terminal multiplexer that actually understands your agents. Oh, and it has incredible mouse support. Bet I can convince you in 6 minutes:

If you're running multiple coding agents (Codex, Claude Code, Pi) at once, you're going to want to watch this. Herdr is a terminal multiplexer that actually understands your agents. Oh, and it has incredible mouse support. Bet I can convince you in 6 minutes:

Jilles Soeters

59,148 views • 2 months ago

Vibe coders and creators can now tap into Pi Network’s massive user base of over 60 million Engaged Pioneers by easily bringing their external AI-created apps to Pi’s real distribution network and utility ecosystem through Pi App Studio! Creators, who may be technical developers or non-technical product people, can build apps using platforms such as Codex, Claude Code, Replit, Cursor, Lovable or other AI-assisted coding tools, and then use Pi App Studio to convert those apps into Pi Apps. That changes the equation for creators. AI can already help people generate apps quickly; the harder part is distribution and turning those apps into products people actually use. Pi helps close that gap. Instead of rebuilding infrastructure from scratch, creators can connect their apps to an existing ecosystem with users, payment capabilities, identity verification, decentralized human infrastructure and platform-level tools already in place. This feature makes Idea → App → Users that much faster. Because your ideas are too good not to be seen. Try the Pi App Studio Now! If you’re a vibe coder or know someone building apps with AI who needs distribution, head to Pi App Studio today to convert AI-created apps into Pi Apps and access the Pi ecosystem’s users, infrastructure, and monetization opportunities.

Vibe coders and creators can now tap into Pi Network’s massive user base of over 60 million Engaged Pioneers by easily bringing their external AI-created apps to Pi’s real distribution network and utility ecosystem through Pi App Studio! Creators, who may be technical developers or non-technical product people, can build apps using platforms such as Codex, Claude Code, Replit, Cursor, Lovable or other AI-assisted coding tools, and then use Pi App Studio to convert those apps into Pi Apps. That changes the equation for creators. AI can already help people generate apps quickly; the harder part is distribution and turning those apps into products people actually use. Pi helps close that gap. Instead of rebuilding infrastructure from scratch, creators can connect their apps to an existing ecosystem with users, payment capabilities, identity verification, decentralized human infrastructure and platform-level tools already in place. This feature makes Idea → App → Users that much faster. Because your ideas are too good not to be seen. Try the Pi App Studio Now! If you’re a vibe coder or know someone building apps with AI who needs distribution, head to Pi App Studio today to convert AI-created apps into Pi Apps and access the Pi ecosystem’s users, infrastructure, and monetization opportunities.

Pi Network

182,099 views • 4 months ago

I rebuilt Posia using the Pi Agent SDK. The core agent is ~15 lines of code, and the harness is so flexible and hackable. Recorded a 17-min walkthrough on: - Pi’s extension system + how to build your own - extensions that cut tool tokens by 80–96% - the core Pi SDK, step by step - building hosted agents with it Adding my Pi extensions setup in AI Builder Club repo here:

I rebuilt Posia using the Pi Agent SDK. The core agent is ~15 lines of code, and the harness is so flexible and hackable. Recorded a 17-min walkthrough on: - Pi’s extension system + how to build your own - extensions that cut tool tokens by 80–96% - the core Pi SDK, step by step - building hosted agents with it Adding my Pi extensions setup in AI Builder Club repo here:

Jason Zhou

66,010 views • 2 months ago

cool thing my Nous Research Hermes Agent does for me daily is collecting things I like here and there, extrapolating context, and saving them meticulously i made a profile just for that! he also built this app so I can always access it without paying Obsidian. what a good boy!

dodo-reach 🦤

19,422 views • 3 months ago

In Pi Network community Pioneers is a very important one. Actually right now all merchants in barter are pioneer! Please don't make pioneers and merchants in a opposite position. They have same interest. The reason many merchant in China provide a lot products to barter is because they want their pi in hands reach GCV and they believe only this price can make Pi successful! If nobody contributes to Pi Eco especially barter how can Eco succeed? They don't support low price is because they understand Pi very well. It will be harvested by crypto whales and it will damage Pi and make pi in dangerous positions if use low price. I believe all pioneers want Pi high price. This is common sense! If you hold a crypto or stock, do you want them high price or low price? You want it low price only when you want to buy ? Is it right! So it can explain anything why some against GCV. However I suggest every one be smart. Now it's time to change to GCV. I believe you hold a lot of pi during close mainnet with different methods. You want more is because you believe pi worth a lot. From now on, no matter what you did before, it's time to unite together to support GCV! Because it is best interest for all Pi Network community! Pi Network #WhatIDoForPi #PiGCV

In Pi Network community Pioneers is a very important one. Actually right now all merchants in barter are pioneer! Please don't make pioneers and merchants in a opposite position. They have same interest. The reason many merchant in China provide a lot products to barter is because they want their pi in hands reach GCV and they believe only this price can make Pi successful! If nobody contributes to Pi Eco especially barter how can Eco succeed? They don't support low price is because they understand Pi very well. It will be harvested by crypto whales and it will damage Pi and make pi in dangerous positions if use low price. I believe all pioneers want Pi high price. This is common sense! If you hold a crypto or stock, do you want them high price or low price? You want it low price only when you want to buy ? Is it right! So it can explain anything why some against GCV. However I suggest every one be smart. Now it's time to change to GCV. I believe you hold a lot of pi during close mainnet with different methods. You want more is because you believe pi worth a lot. From now on, no matter what you did before, it's time to unite together to support GCV! Because it is best interest for all Pi Network community! Pi Network #WhatIDoForPi #PiGCV

Doris Yin 东方紫莲🪷

39,074 views • 3 years ago