[0:00] One AI agent, one task at a time. That's
[0:02] the ceiling, and it's slow. So, we break
[0:04] it with Hermes. Run a whole team of
[0:06] agents in parallel. So, you take one
[0:08] job, split it into pieces, and put one
[0:11] agent on each, all working at the exact
[0:13] same time. Same role, different target.
[0:16] Five agents, five companies all at once.
[0:19] And you're never capped at five. Need a
[0:21] 10th next week? Text the bot one line,
[0:23] and it's in. No setup, no new agents.
[0:26] Then one last agent reads everything
[0:28] they found and turns it into a single
[0:30] report. And the whole thing is
[0:32] automated. It runs every morning on its
[0:34] own before you're even up. This is
[0:37] what's on my phone when I wake up, a
[0:39] full professional report. As you can
[0:41] see, five competitors built and
[0:43] delivered overnight. I didn't touch it.
[0:46] No code.
[0:47] Let me show you how to build this from
[0:49] scratch.
[0:54] So, we've got Kanban open.
[0:56] We're building four agents that research
[0:59] for companies at the same time. Then one
[1:01] that turns it all into a report. So,
[1:03] first, let's click on profiles,
[1:06] where you set up your agents.
[1:08] So, here's how I did it the first time.
[1:10] Four agents, Asana, ClickUp, Monday, and
[1:14] Notion. Four companies, four agents. And
[1:16] that's wrong because an agent is a role,
[1:19] not a topic. A researcher is a
[1:22] researcher whether it reads Asana or
[1:24] Notion. Same job, you're just pointing
[1:27] it at a different company.
[1:29] It's like hiring four people because
[1:31] you've got four reports to write. You
[1:33] don't. You hire one researcher and hand
[1:36] them all four. And one researcher
[1:39] scales. 10 competitors next week? Same
[1:42] agent, more tasks. If you have four
[1:44] profiles, you'd build a new one every
[1:47] single time, and it's a dead end. So,
[1:49] this go. Let's delete them.
[1:53] And of course, we keep default. And now,
[1:55] the right way. One profile. Let's click
[1:58] on create.
[1:59] Let's name it
[2:01] researcher.
[2:02] For the description, let's keep it
[2:04] simple.
[2:05] Researches a company's pricing and
[2:08] features. And that's the role stated
[2:11] once. Doesn't matter which company.
[2:13] That's the whole point. The model is
[2:16] DeepSeek-V4-Flash,
[2:17] not the pro one, because these agents
[2:20] just fetch and summarize. Flash is fast,
[2:23] cheap, plenty for reading a pricing
[2:26] page. The smart model comes in later,
[2:29] where it earns it.
[2:30] So, DeepSeek-V4-Flash and create. So, we
[2:34] have one researcher, and that's the
[2:36] agent doing all four jobs.
[2:39] So, one more thing before we built.
[2:41] Right now, the default agent runs the
[2:43] bot, but it can't create tasks on its
[2:46] own from a Telegram message, and we need
[2:49] that at the end. So, we add the tool
[2:51] now. Let's open our terminal, and let's
[2:53] copy this command. So, this backs up the
[2:56] config file, a copy, in case we break
[2:59] something editing it. Anything goes
[3:01] wrong, we restore from this. So, let's
[3:03] run it. No output means it worked.
[3:07] Okay, great. Our next command. Let's
[3:10] paste it. So, this one adds Kanban to
[3:13] the tool set list. The line that gives
[3:15] the agent the power to create tasks.
[3:17] Let's run it.
[3:19] Great. Now, let's check.
[3:22] And here we go. We added Kanban. So, we
[3:25] have Hermes-CLI
[3:27] and Kanban, both there. Now, let's
[3:30] restart the gateway.
[3:32] This will reboot the agent, so it loads
[3:34] that new Kanban tool. Let's run it.
[3:38] Okay.
[3:39] Gateway is shutting down.
[3:41] So, now we can see that start has
[3:44] failed. Let's fix that. First, let's
[3:48] type anything to Telegram, and it's
[3:51] working.
[3:52] But, we still have start failed. Let's
[3:54] click on it.
[3:56] Okay. So, the easiest way to fix this is
[3:59] to copy this.
[4:02] I know it sounds very lazy, but let's go
[4:04] to chat.
[4:06] Let's paste this here and tell Hermes to
[4:09] fix it.
[4:11] And we'll wait.
[4:15] Ah, here we go.
[4:16] It's running right now. I think it's
[4:19] fixed. It's usually because it's a
[4:21] Telegram conflict, two gateways fighting
[4:24] over the same bot.
[4:25] So, by telling Hermes to fix it, it will
[4:28] close all the gateways and keep defaults
[4:30] running.
[4:34] Ah, here we go. Everything is fixed. You
[4:37] can see the root cause. The researcher
[4:39] profile's gateway was holding the
[4:41] Telegram bot token, blocking the default
[4:44] profile's gateway from starting. For
[4:46] this kind of problems, it's better to
[4:48] tell Hermes to fix it itself.
[4:50] So, [snorts] setup's done. Now, we pawn
[4:52] this one agent at four companies and let
[4:55] it run. So, let's head over to Kanban.
[4:58] Our dashboard. So, first thing to do is
[5:01] to create the tasks, one per company.
[5:04] So, first one is ClickUp.
[5:08] Let's create, and then we have Monday.
[5:13] Then Asana.
[5:16] And lastly, Notion.
[5:20] Okay. Four tasks, empty for now, no
[5:23] assignee, no description, nothing. They
[5:25] just sit there.
[5:27] So, now the one that matters, which is
[5:29] task five, the agent that pulls
[5:31] everything together.
[5:33] So, let's add it.
[5:35] The name is the synthesizer. Let's
[5:37] assign it to default.
[5:41] And for parents, all of these tasks are
[5:44] its parents. So, let's select one for
[5:47] now. Let's create and let's open it up
[5:51] because this one needs real prompt.
[5:53] First thing's first, let's add more
[5:56] parents, which is Monday,
[5:59] Asana, and Notion.
[6:01] And for the description, let me copy my
[6:04] prompt and let's explain it.
[6:07] So, this is the prompt that matters and
[6:10] I want you to notice how it's written.
[6:12] I'm not telling it which buttons to
[6:14] press. I'm giving it the goal and the
[6:16] output. Read what the four researchers
[6:18] found, build a report, here's what the
[6:21] report should look like. The design,
[6:23] that's my brand colors, dark background,
[6:27] gold fonts. I'm adding it a style guide.
[6:30] The content, a cover, a takeaway, line
[6:33] up top so it makes a point instead of
[6:36] just dumping data, charts, a comparison
[6:39] table, top five features. And the last
[6:41] line, the media one, that's what drops
[6:44] the finished file straight onto my
[6:46] phone. What I'm not doing is telling it
[6:48] how to install anything or which library
[6:51] to use. I give it the what, it figures
[6:53] out the how. That's what makes an agent
[6:56] reliable instead of brittle. Okay? So,
[6:59] we are done here. Let's save and let's
[7:02] go back to our dashboard. And watch,
[7:05] even with default assigned, the
[7:07] Synthesizer does not launch. It just
[7:09] sits because it's waiting on all four
[7:11] parents to finish first. That's the
[7:14] funneling. One agent holding until
[7:16] everything it needs is done. Now, let's
[7:19] arm the researchers. Each one gets the
[7:22] same agent, researcher, and the same
[7:24] instructions pointed at a different
[7:27] company. So, let's start with ClickUp.
[7:30] So, assignee is researcher
[7:33] and let's edit the description. Save.
[7:38] Second one is Monday, so researcher.
[7:41] Let's edit it.
[7:45] So, for ClickUp here, after
[7:47] researcher, let's hit enter. Now, it's
[7:50] done.
[7:51] Asana
[7:52] same thing
[7:55] Let's edit the description.
[7:58] And Notion So, researcher
[8:02] And let's paste our command, our prompt,
[8:06] and click save. Now, they start on their
[8:08] own in about a minute.
[8:11] The bot checks every 60 seconds, but I
[8:13] don't want to stare at it. So, let's
[8:16] click on nudge. And here we go. They're
[8:18] in progress. So, the first one is
[8:21] already done. Let's wait for the other
[8:23] three.
[8:24] So, while these run, here's the thing
[8:26] people miss about agents. They're not
[8:28] faster because they type faster, they're
[8:30] faster because they work at the same
[8:32] time. One of me does these four
[8:34] back-to-back. Four agents do them at
[8:37] once.
[8:38] And they're finishing.
[8:40] So, all four done, and now task five
[8:42] should wake up. Let's scroll here.
[8:47] So, it's in ready. Let's click on nudge.
[8:50] And now it's in progress, and let's wait
[8:53] for it.
[8:54] It will be reading all four reports and
[8:56] building the PDF.
[8:58] >> [snorts]
[8:58] >> First run, it installs the rendering
[9:00] library, so it takes a minute.
[9:03] And then the agent will solve its own
[9:05] setup.
[9:06] So, here, nothing for me to do. We'll
[9:08] wait for the Telegram message.
[9:11] And there it is, straight to Telegram. I
[9:14] didn't write this, didn't design a
[9:16] single page for agent research. One
[9:19] agent built it, and it landed on my
[9:21] phone as a finished file.
[9:23] So, let's open it.
[9:25] And look at this. A cover page, title,
[9:28] the four platforms, the quarter, the
[9:31] date. This isn't a text dump. It's a
[9:33] document. And right here, page two, the
[9:36] executive summary.
[9:38] Read that first line, ClickUp wins on
[9:40] price-to-feature ratio.
[9:42] That's not data, that's a verdict. The
[9:45] agent didn't just collect numbers, it
[9:47] took a position.
[9:48] I come back to that because that's the
[9:50] most important thing in the whole
[9:52] report. And the chart, two of them,
[9:54] entry tier and most popular tier, side
[9:57] by side in my colors, the gold, the dark
[10:00] background, it built those from the
[10:02] numbers the researchers built. Then we
[10:05] have the full breakdown, every tier,
[10:08] every price for all four, ClickUp,
[10:10] Asana, monday, Notion, what each plan
[10:14] costs and what you actually get. And you
[10:17] have here top five features across
[10:19] platform ranked, AI agents, automations,
[10:22] the view types, integrations, enterprise
[10:25] security one to five. On the last page,
[10:28] this is the one, a positioning matrix.
[10:31] Each platform, its price range, its
[10:34] wrist strength, what it's for, ClickUp
[10:36] for cost-conscious teams, Asana for
[10:39] regulated enterprises. This is strategy,
[10:42] not a spreadsheet. So, for agents
[10:44] research this, something you'd actually
[10:47] send a client. But there is one more
[10:49] thing I still had to do. Sit here and
[10:52] start it. So, let's fix that. Let's make
[10:54] it an automated workflow where you can
[10:57] add competitors and set that to work
[11:00] every day.
[11:02] So, to do that, let's go first to
[11:04] Telegram. So, the first thing to do is
[11:07] to give the bot a memory, my competitors
[11:09] once right here in the chat. So, let's
[11:12] paste this command. Okay, memory my
[11:15] competitor list, ClickUp, monday, Asana,
[11:17] Notion. Remember this, I'll ask you to
[11:19] run research on them later. Let's run
[11:21] it.
[11:24] And now it knows my list and I can
[11:26] change it anytime just by talking to it.
[11:29] You'll see. So, one more thing in its
[11:31] memory, the builder's instructions. So,
[11:34] when it runs, it knows exactly how to
[11:36] assemble the report. This is the brain
[11:38] behind that PDF. Let's paste this, and
[11:42] let's run it.
[11:46] So, now we have the list and the builder
[11:48] both in memory. Now, the schedule. I
[11:50] want it to ping me every morning and ask
[11:53] if I want to run. And here's the catch,
[11:55] a scheduled job can't ask me and then
[11:58] sit there waiting for an answer. It runs
[12:00] alone, no chat. So, it pings me and I
[12:03] reply separately. Ping, then reply. So,
[12:07] let's paste our cron command, and let's
[12:10] run it. And that's the whole engine in
[12:13] one instruction. The ping only asks, but
[12:16] go builds everything. A research task
[12:18] per company and the builder on top
[12:21] waiting on all of them. And I made it
[12:23] confirm the count back to me, so I
[12:25] always know the builder's there. So, I'm
[12:28] not waiting until 9:00 a.m. tomorrow to
[12:29] show you. I'll tell it to run the job
[12:32] right now. So, let's say, "Run the daily
[12:35] search ping job now." And let's run it.
[12:39] So, let's wait for it. It's running, and
[12:43] there's the ping, "Morning. Here's your
[12:45] list ClickUp Monday Asana Notion.
[12:48] Edit or say go." And it pulled that from
[12:51] memory. I didn't paste it. So, let's
[12:53] say, "Go." And then enter.
[12:56] And it confirms. So, it's creating the
[12:59] four tasks, as you can see here. And
[13:02] then the synthetizer. So, let's go to
[13:05] the board. Here you go, four researchers
[13:07] and one synthetizer. Great.
[13:11] So, now let's wait for our PDF, and
[13:13] after that we run a new competitor.
[13:17] Great. So, everything is working. Now,
[13:20] the part that makes this useful, the
[13:23] dynamic workflow, because our list isn't
[13:26] locked. When a new competitor shows up,
[13:28] we just tell it to add it. For example,
[13:32] let's add Linear to the list.
[13:34] So, add Linear to my competitor list
[13:37] just update the memory list. Do not
[13:39] create or run anything yet. Let's run
[13:41] it.
[13:43] And it just updates the memory. Nothing
[13:45] runs. Adding a name and running the
[13:47] research are two different things. It
[13:50] knows the difference. Lists, we have
[13:52] five competitors now.
[13:54] And the board hasn't moved.
[13:56] Now, let's say go.
[13:59] And hit enter.
[14:01] So, five research tasks now and the
[14:03] builder on top of all five. So, I have
[14:06] six.
[14:09] So, let's see the board. So, we have
[14:12] five research tasks and the synthesizer,
[14:15] the sixth one.
[14:17] So, five researchers, one for each
[14:19] company including Linear, and the
[14:21] builder waiting on all five. Great.
[14:24] So, I added one word to a chat message
[14:26] and the whole machine reshaped around
[14:28] it.
[14:29] That's an agent, not a script. I didn't
[14:32] rewire anything. I told it.
[14:34] So, here is where we landed. Every
[14:36] morning at 9:00, it pings me with my
[14:38] list. I reply, edit it, or just say go.
[14:42] And it builds the whole thing. Finds
[14:44] out, researches everything, sends me the
[14:46] finished report.
[14:49] I set it up once. Now, it runs itself.
[14:55] And here's the final report. Same
[14:57] engine, but now with Linear in it. Let's
[15:00] click on it.
[15:02] And right on the cover, we have ClickUp,
[15:04] Asana, Monday, Notion, and Linear.
[15:07] If you scroll down,
[15:09] we can see Linear here as part of the
[15:11] chart. So, it didn't fill a slot. It
[15:13] researched Linear and took a position on
[15:16] it. That's the whole thing. Five
[15:18] competitors, one word added to a chat,
[15:21] four reports on my phone, and tomorrow
[15:23] at 9:00 it asks me again.
[15:26] And that's parallel agents in Hermes.
[15:28] We're all many targets running on their
[15:31] own every morning. If you want to run
[15:33] Hermes 24/7, there is a VPS link in the
[15:36] description. Use the code nocodehack for
[15:39] a discount. Like the video, subscribe
[15:41] for more Hermes builds, and I'll see you
[15:43] in the next one.