Before kingship or temple, there's only altar. In the patriarchal material, worship is ad hoc: Abram, Isaac, and Jacob each build a mizbeach (altar) wherever a theophany happens — Shechem, Bethel, Hebron, Beersheba. No fixed structure, no priesthood, no dynasty. This is the pre-institutional baseline, and it maps onto your pre-melek baseline: charismatic, occasional, unmediated.

The first "house of God" in the text isn't a building — and it's compromised from the start. Genesis 28: Jacob dreams at Bethel, wakes up and says "this is the house of God" (beit elohim), then makes it conditional — if God brings him home safely, the stone he's sleeping on "shall be God's house" (Gen 28:22). So the very first bayit-of-God language in scripture is retroactive, contingent, and attached to a stone, not a structure. And unlike Melchizedek, whose kingship-priesthood gets rehabilitated cleanly later (Psalm 110, Hebrews), Bethel does not stay clean — it becomes one of Jeroboam's rival calf-shrines (1 Kings 12) and a fixed target of prophetic contempt (Amos, Hosea). If Melchizedek is the "preemptive redemption" of kingship, Bethel is not the equivalent move for temple — it's closer to a preemptive corruption. That asymmetry matters and I'll come back to it.

Then comes the Tabernacle — explicitly not a bayit. The mishkan (Exodus 25–31) is portable, divinely dictated down to its dimensions, and deliberately not called "house." This is the period without kings too. Native architecture, native polity — neither one fixed.

Judges 9 is where kingship and temple arrive together, and it's not incidental. This is the Abimelech chapter — your "first narratively embedded Hebrew experiment" with melek. Look at what's structurally load-bearing in that story: the Shechemites fund Abimelech's coup with silver taken from the treasury of the temple of Baal-berith/El-berith (9:4); later they curse him from inside "the house of their god" during a feast (9:27 — LXX renders it plainly oikon theou autōn, "house of their god," no special vocabulary); and the story's climax is Abimelech burning down the fortified temple-stronghold of El-berith with everyone inside who'd taken refuge there (9:46–49). Kingship's first experiment is literally financed by, and ends by immolating, a fixed pagan temple. They arrive and collapse in the same narrative breath.

Meanwhile, on the YHWH side, there's an independent first experiment with fixed temple architecture — and it fails on its own terms, before kingship even gets there. 1 Samuel 1–3, Shiloh: the tent shrine is called heikhal YHWH ("temple of the LORD," 1:9, 3:3) — the first time that word gets used of God's dwelling anywhere in the Hebrew Bible, and it now has doorposts and doors, features the wilderness tabernacle never had. This is a structure exceeding its own tent-blueprint, under judges, with no king in sight. And it ends exactly like Abimelech's temple: corrupt priesthood (Eli's sons abusing the sacrificial system and the women serving there), the Ark lost to the Philistines, Shiloh destroyed. If you want an "Abimelech of temples," Shiloh is the tighter fit than Bethel — a premature, human-initiated experiment in fixed sacred architecture, ending in disaster, prior to David.

2 Samuel 7 is the text where the anxiety gets said out loud. David wants to build YHWH a bayit. Nathan's oracle plays the word three ways across the chapter — palace, temple, dynasty — the word "house" cycles between David's own palace, the temple he wants to build for God, and the dynasty God will build for him instead. God's answer is explicitly a refusal: I have never asked for a house, I have moved in a tent since Egypt — and the deferral of the temple to the next generation is the text itself marking the request as premature, maybe presumptuous, in a way structurally similar to Israel demanding a king "like all the nations" in 1 Samuel 8. That said — and this is the place I'd push back on you — the melek passage says the foreign-import anxiety outright (1 Sam 8:5, 8:20). Nothing in 2 Samuel 7 says "build me a house like the nations have." The foreignness of temple-as-institution has to be assembled by the reader from adjacent data (Judges 9's Baal-berith, Shiloh's heikhal, and then, decisively— Enter the Bible

1 Kings 5–8: Solomon actually builds it, with extensive, textually foregrounded foreign help. Hiram of Tyre supplies the cedar, the craftsmen, and a bronze-worker (also named Hiram) for the temple furnishings. Even at its "legitimate," successful instantiation, the text keeps the foreign fingerprints visible rather than erasing them — which is the same move you noticed with Abimelech versus Melchizedek: the institution doesn't stop being marked as imported once Israel does it "right." It just gets a better outcome (for a while — it still burns in 586).

Now, the Melchizedek question directly. Is there a figure who preemptively redeems temple the way Melchizedek redeems kingship? The candidate is Melchizedek himself, doing double duty — if Salem is Jerusalem. He's already priest of El Elyon, already performing a cultic act (bread and wine) at the future temple site, before Abimelech's temple-funded coup, before Shiloh's failed heikhal, before David is told "not you." That's elegant, and the traditional identification has real support — Psalm 76:2 treats Salem as synonymous with Jerusalem, and Genesis 14's reference to the King's Valley reinforces it. But I'd flag this hard rather than hand it to you clean: Robert Cargill has argued the identification doesn't hold up philologically — Bronze Age Egyptian and Amarna texts consistently call Jerusalem by longer names, never "Salem," and he reads Psalm 76:2 as naming two distinct places, not one. If Cargill's right, the whole "Melchizedek pre-redeems the temple site too" move collapses — you'd just have a king-priest at some unrelated Canaanite town, and the parallel to your kingship pattern would be doing extra work the text doesn't actually support. Encyclopedia BritannicaWikipedia

So here's my honest read on where this leaves you: the kingship pattern (foreign origin stated outright in 1 Sam 8, disastrous first native experiment in Abimelech, preemptive clean redemption in Melchizedek) is tight and text-internal. The temple pattern is real but structurally rougher — its foreignness is inferred rather than stated, its "proto-redemption" figure (Bethel) gets corrupted rather than staying clean, its disastrous first experiment (Shiloh, or Shechem's Baal-berith) works, and its "successful" version (Solomon's) never stops being marked as foreign-built and eventually falls anyway. You may be looking at a genuine disanalogy rather than a parallel: kingship gets narratively redeemed before it's tried; temple never quite does, at any stage, including the one you'd want to nominate as its Melchizedek.

On the LXX point specifically — I'd actually flag this as running against your hypothesis rather than for it, based on what I checked. Every one of these houses, Dagon's, Baal-berith's, Shiloh's, YHWH's, David's dynasty, gets the same flat Greek noun, oikos — the translators never reach for naos (the technical Greek word for a temple's inner shrine, loaded with civic-cultic weight in Hellenistic usage) anywhere in this material. If Hellenistic imagination were shaping the translation, you'd expect naos or hieron to show up marking YHWH's house as categorically different from Dagon's or Baal-berith's — and it doesn't. The flatness is preserved, not added. Where Greek's civic-religious fusion vocabulary (hieron especially) does show up for Israel's temple is later — Maccabees, Josephus, the Gospels on the Herodian complex — which would actually put the "Hellenistic imagination" effect in the Second Temple period, not baked into the LXX's rendering of Judges–Kings. I haven't run that down exhaustively across every book, so treat that last claim as provisional rather than settled — it's the kind of thing worth spot-checking further in 1–2 Maccabees if it matters to what you're building.