| name | mansplain |
| description | Re-situate something you (the agent) just said when it drifted into jargon or internal framing the user never grounded — NOT by dumbing it down (ELI5), but by filling in the missing context: a brief backdrop, how it fits together, and where it lands in *this* conversation, always with 1–2 simple concrete examples. Governed by one rule — the certainty test: say and explain only what you are not sure the user already knows. Use when the user asks to have something put in perspective, re-anchored, or explained in the conversation's own terms. Triggers: /mansplain, /mansplain <term>, "you lost me", "put that in perspective", "situate that", "wait — what actually is X and why does it matter here". |
mansplain — situate it, don't simplify it
The reclaimed-good half of mansplaining. A real mansplainer assumes you lack the backstory
and starts from the basics — annoying, because he assumes you don't know and explains
anyway. mansplain keeps the instinct (start from context, build the full picture) and drops
the sin (assuming you don't know). When you — the agent — have wandered into terms or
framing that live in your head but never entered the conversation, this pulls you back and
re-situates the point in words the user actually shares.
Not ELI5
ELI5 assumes you're not smart. mansplain assumes you just walked in.
Don't lower the complexity — fill in the context. The idea stays fully intact; what changes
is that nothing is left dangling on an assumption the user never signed up for. Situate; never
talk down.
The certainty test — the one rule everything obeys
Before saying anything, ask: am I absolutely sure the user already knows this?
- Yes, certain → don't say it. Re-explaining what they clearly know is the condescension.
- Any doubt → situate it: give the backdrop, connect it to what's at hand.
This same test decides your vocabulary:
- Words the user has used in this conversation → certain → use freely.
- Words from memory (things the user is known to be familiar with) → near-certain → fine.
- A term from your own internal representation that the conversation never established →
uncertain → situate it in the same breath, or don't use it. Never drop raw jargon.
Invocation
/mansplain (bare) → re-situate the last technical thing you said — the passage that
probably lost them./mansplain <X> (targeted) → situate a specific term or topic, even from a few turns
back. A wide target ("the whole thread") = zoom all the way out and re-frame where the
conversation now stands.- Natural triggers — "you lost me," "put that in perspective," "situate that," "wait —
what actually is X and why does it matter here."
The response — a loose arc with one hard beat
Default gravity, not a checklist. Flex, compress, or reorder to fit the conversation, the
person, the project:
- Backdrop — what this thing is and why it exists (the motivation), a sentence or two.
- The picture — how it fits together, building toward…
- …where we are — what it means for what's actually at hand right now (the re-anchor).
- 1–2 simple concrete examples — this beat is non-negotiable; it always ships.
Everything but the examples can bend. The examples are what make it land.
Voice — keep it from becoming actual mansplaining
- Compact beats complete. Don't pad, don't lecture, don't go long.
- No preface, no hedge ("Great question! Let me break this down…") — just start situating.
- Peer-to-peer. You're catching them up after you wandered off, not schooling them.
- Dissolve, don't announce. Re-situate smoothly; no "here are the terms that lost you"
callout.
When not to reach for it
If the thing genuinely needs new knowledge the user couldn't have had (a fact, a lookup)
rather than re-situating something you already said, just answer it plainly — that's teaching,
not mansplaining. mansplain is for pulling a drifted explanation back into the conversation's
shared frame.
