Not who you love. Who you were actually in contact with — the recurring calls, the group chat, the collaborator, the client you dread, the friend whose news is always bad.
Pashupati's claim about that list is uncomfortable and short:
If you are of the highest energy quality, then the people in your life will be of the highest energy quality. But if you are failing all the time and losing energy, then you'll have wavering people around you — in the mood today, not in the mood today, I don't feel like it. Who wants to be surrounded by such people?1
It's a readout claim. The circle isn't primarily something you chose; it's something that assembled around the state you've been in, and it therefore reports on that state more accurately than your own assessment does.
He gives it without softening, in the middle of the adbhuta material, as a prerequisite for having any energy at all:
You have to stop leaking energy. You have to find your dharma, you have to go along with it, and you have to do things which make your heart happy. Get rid of everybody who makes you unhappy in your life.2
And then the pre-empt, aimed at the obvious objection:
You really have to. You can't have people in your life making you unhappy and then say I'll be happy by myself, it doesn't matter about my environment, I am — shut up, man. Don't fool yourself. People in your life matter.3
That's aimed squarely at a particular consolation: the idea that inner work makes you independent of your environment. His position is that it doesn't, that the claim is self-deception, and that the environment is doing continuous work on you whether or not you've decided it shouldn't.
The vocabulary is worth taking seriously because it's precise about the failure mode.
A leak isn't a cost. A cost is paid once, visibly, and you can decide whether it's worth it. A leak is continuous, low-grade, and undetected — the tank is emptying at a rate you never measure, and you attribute the emptiness to something else.
Which matches how draining relationships actually operate. Nobody has a single catastrophic interaction and identifies it as the problem. What happens is that a standing arrangement — the weekly call, the group chat, the collaborator who needs managing — takes a small amount continuously, and the deficit shows up as I don't seem to have any energy for the work lately.
The reason this is a creative-practice page rather than life advice: everything else in this framework says output is downstream of state. If the state is being drained by infrastructure nobody is auditing, then the intervention isn't better working habits. It's the audit.
The page should be honest about what it's holding, because the instruction is unusually blunt and the failure modes are serious.
Get rid of everybody who makes you unhappy is, stated flatly, a licence to discard people during difficult periods — including people whose difficulty is temporary, or whose difficulty is a claim on you that you have obligations toward.
Nothing in his framing distinguishes:
A genuine drain — a standing arrangement that takes continuously and returns nothing, sustained by habit or guilt.
A hard season — someone you care about going through something, which is expensive and is what relationships are for.
Your own state — a period in which everyone seems draining, which is a fact about you rather than about them.
The third is the most dangerous omission, because the framework's own logic predicts it. If a depleted person's circle assembles around their depletion, then a depleted person's perception of their circle is also depleted — and acting decisively on that perception is how people burn relationships they'll want back.
[FLAG] — document, don't endorse. The observation about infrastructure is real. The instruction as stated has no guard on it at all.
Two of the three items in his instruction get less attention than the removal, and they're doing separate work.
Find your dharma and go along with it. In this framework dharma is what vīra grows out of — the thing you've taken a side on, the position you'd defend. So the claim is that operating against your own position leaks continuously. The client work that contradicts what you say publicly, the format you keep making because it performs, the topic you cover and don't believe in — each of those requires a small ongoing act of self-management, and the management is the leak.
Do things which make your heart happy. Which sounds like a platitude and, inside this framework, is a structural claim: since the rasas sit at the high end of the emotional band, and since the whole system says output is downstream of state, activities that raise the band aren't recreation. They're supply.
Put with the third item, the instruction has a shape it lacks when only the removal is quoted: stop the outflow, align the work with the position, and increase the input. Three levers, and removal is the one everybody reaches for first because it's the only one that feels decisive.
It's also the slowest to show a result and the easiest to get wrong, which is a reasonable argument for doing it last.
You've had six weeks where nothing has come out and you've tried everything at the level of the work.
New schedule. Earlier start. Different room. None of it moved.
So you do the audit at the other level, and you do it on paper because it needs to be counted rather than felt.
Every recurring contact of the last month. Not events — standing arrangements. Nine of them.
Then one mark against each: after this, do I have more or less than before?
Six neutral. Two clearly more — one of them a call you nearly cancelled twice.
And one that's unambiguous. The Thursday thing. Ninety minutes, every week, for eleven months, and you can't remember the last time you finished it without needing an hour to recover. You've been treating it as a fixed feature of the week.
Now the check that stops this becoming a cull, and it's the one the source omits. Is this a drain or a season? You go back: was it like this in January? And it was. Eleven months, consistent. That's a drain, not a hard patch.
Had the answer been no, only since March, the correct move would have been the opposite — because that's a person in something, and leaving is what this framework's own karuṇā material spends three hours arguing against.
So you change one thing. Not a confrontation, not an ending — Thursday becomes monthly.
Four weeks later you've published twice, and you'd have sworn the problem was your morning routine.
The best evidence in the source is a case where someone did this deliberately and in the positive direction.
When Reika decided to run her retreat, Pashupati's account of her first move isn't about marketing. It's: who are all the high-energy people I know? Let's involve them.4
She recruited AJ in Dubai, who built the website. She pulled Pashupati in, and Pratish. And by his account, only high-energy people ended up registering.5
[UNVERIFIED] throughout and 🚩 the teacher is in the story.
But the structure is the interesting part, and it's the inverse of the audit above. She didn't remove drains — she assembled deliberately, treating the circle as something to be built for a purpose rather than inherited.
That's the more usable half of the doctrine and he never separates it. Removal is fraught, slow, and full of the failure modes flagged above. Addition is neither: you can put someone in your week without taking anyone out, and the effect on the readout is immediate.
Which suggests the practical order is the reverse of the instruction. Add before you subtract, and a fair amount of what looked like a drain stops registering as one once the ratio changes.
If removal is fraught and addition is not, then the usable form of this doctrine is narrower than the instruction and considerably safer.
Audit, don't act. Simply counting the standing arrangements and marking each one more-or-less takes twenty minutes and changes nothing about your relationships. Most of its value is in the counting — people are consistently surprised by which arrangement turns out to be the expensive one, and the surprise is the finding.
Adjust frequency before existence. Weekly to monthly is available, reversible, and requires no conversation. It captures most of the effect of removal with almost none of the cost, and it's the move his own framework should prefer given everything it says about karuṇā.
Add first. The Reika case is the addition version, and it's the half of the doctrine with no failure mode attached.
What's left after those three is a small residue of genuine, sustained, unreciprocated drains — and by the time you've done the other three, you'll have much better evidence about which ones those actually are, because you'll have watched them across a period when your own state was improving.
Which is the ordering the source doesn't give and the framework's own logic requires: fix your state first, then re-read the circle, because a depleted person's list is not trustworthy.
Evidence. The infrastructure claim is sound and unexotic: standing arrangements consume continuously, and continuous small costs are the ones that go unaudited. That's checkable in an afternoon.
Tension — no guard. The instruction cannot distinguish drain from season from your own state, and the framework's own logic predicts the third case will produce false positives.
Tension — the karuṇā contradiction. Three hours arguing that care is the base rasa of all action, and then get rid of everybody who makes you unhappy. He doesn't notice.
Open question. Is a drain ever an obligation? The framework has no category for a relationship that costs you and that you should keep anyway, and most people have several.
Set this against Karuṇā as Care — his own — and the two pages disagree.
The karuṇā material's whole argument is that care for specific people is the base rasa of the entire action cluster, that it must run inward and outward, and that its loss produces the ascetic. Care is the thing you build everything on.
This page says remove people who make you unhappy.
Those aren't reconcilable at face value, and the best available resolution is one he doesn't state: karuṇā is directed — you care about a particular audience, a particular person, the one you'd help unpaid — and it is not a general obligation to everyone in contact with you. The circle audit is about unchosen standing arrangements; karuṇā is about a chosen orientation.
That's coherent, and it's also exactly the reasoning a person uses when they're about to do something they'll regret. The vault records the tension rather than smoothing it.
Against Be Aesthetic, Not Ascetic — also his — the tension runs the other way. That page says withdrawal from people is depletion wearing a costume. This one prescribes removing people as a cure for depletion. Same corpus, opposite directions, and the difference is presumably meant to be which people — which is precisely the judgement the framework supplies no help with.
To Somatic Trauma Theory Hub — psychology
The somatic and attachment literature has a well-developed account of co-regulation: nervous systems entrain to each other, and the states of the people you're regularly in contact with are not metaphorically but physiologically transmitted. Regulated company regulates; dysregulated company dysregulates, and neither party has to say anything for it to occur.
Which is stop leaking energy, with a mechanism attached.
What the pairing produces is a serious correction to the prescription. The clinical account says co-regulation is bidirectional and non-optional — you are also a source for everyone in your circle, which means removal isn't a neutral act of self-care but a withdrawal of regulation from someone else. And it says the effect is state-dependent rather than a property of persons: the same relationship regulates in one period and dysregulates in another, depending on where both parties are. That is exactly the drain/season distinction the source lacks, and the literature supplies both the reason it matters and the reason a depleted person is the least able to make it.
To Creator Economy Hub — business
The creator-economy corpus treats environment tactically — the mastermind, the peer group, the room-you're-in advice, the standard finding that a creator's trajectory tracks their circle's ambition level. It's usually framed as opportunity access: better rooms produce better introductions.
This says the mechanism is upstream of opportunity and operates continuously.
The insight the pairing produces concerns what the room is actually supplying. The opportunity account predicts that the value of a peer group is realised in discrete events — an introduction, a referral, a collaboration — and that a group producing none of those is worthless. The energy account predicts the value is in the baseline, delivered continuously, and largely invisible: you work differently all week because of who you spoke to on Tuesday. Those recommend different behaviour. Opportunity says optimise for who can help you; energy says optimise for who leaves you able to work — and the two lists overlap far less than anyone expects, because the most connected person in a field is frequently the most depleting.
Sharpest implication. The circle is a readout of your state, not just an influence on it — which means a depleted person's assessment of who is draining them is itself produced by the depletion. The framework's own logic predicts false positives at exactly the moment it recommends decisive action, and it supplies no guard.
Generative questions.