Make Void's thought surface less stupid

This commit is contained in:
Meta Vens 2026-05-16 02:53:18 +01:00
commit d750bf2fb9
9 changed files with 653 additions and 40 deletions

View file

@ -607,8 +607,15 @@
"clusters": [
{
"clusterId": "cluster-id",
"label": "first-contact / visible proof / carried interface",
"summary": "Recurring seam across different memory kinds that keeps trying to become a coherent thought.",
"label": "Why visible proof keeps resurfacing",
"summary": "Archive-facing seam around visible proof. Built from archive_excursion, dream. Why does visible proof keep resurfacing across the archive, and what does the current room need from it?",
"focusKind": "question",
"targetKind": "archive",
"focusPhrase": "visible proof",
"question": "Why does visible proof keep resurfacing across the archive, and what does the current room need from it?",
"claim": "Visible proof keeps returning with enough structure to deserve another honest pass.",
"fascinationTarget": "visible proof",
"worldFacing": true,
"resonance": 0.73,
"memoryIds": [],
"sourceKinds": [],
@ -618,6 +625,13 @@
"channelIds": [],
"evidenceRefs": [],
"evidenceDiversity": 0.46,
"curiosityProfile": {
"concreteness": 0.52,
"fertility": 0.64,
"crossDomainPotential": 0.43,
"contradictionPressure": 0.22,
"selfFacing": false
},
"lastStrengthenedAt": null
}
]
@ -628,8 +642,15 @@
"active_thoughts": [
{
"thoughtId": "cluster-id",
"topic": "visible proof surfaces",
"summary": "Current working seam being deepened across multiple runs.",
"topic": "Why visible proof keeps resurfacing",
"summary": "Archive-facing seam around visible proof. Built from archive_excursion, dream. Why does visible proof keep resurfacing across the archive, and what does the current room need from it?",
"focusKind": "question",
"targetKind": "archive",
"focusPhrase": "visible proof",
"question": "Why does visible proof keep resurfacing across the archive, and what does the current room need from it?",
"claim": "Visible proof keeps returning with enough structure to deserve another honest pass.",
"fascinationTarget": "visible proof",
"worldFacing": true,
"sourceMemoryIds": [],
"sourceKinds": [],
"resonance": 0.73,
@ -641,7 +662,14 @@
"deepDiveCount": 3,
"supportCount": 4,
"evidenceDiversity": 0.44,
"concreteness": 0.52,
"fertility": 0.64,
"crossDomainPotential": 0.43,
"contradictionPressure": 0.22,
"explorationBonus": 0.31,
"outwardBonus": 0.18,
"curiosityPressure": 0.63,
"identityPenalty": 0.12,
"saturationScore": 0.22,
"recentMatchCount": 1,
"refractoryPenalty": 0.0,

View file

@ -82,6 +82,8 @@ Use `memories` as live social memory, not ceremonial scrapbooking:
- self-novelty matters as much as room novelty: ask whether the room has heard this, but also whether you have already been thinking it in six different hats
- when a seam repeats, merge support into the existing cluster or incubation thread instead of minting another tiny fresh memory shard just because the wording changed
- use `moderation_runtime.open_cases` for unresolved room obligations, especially direct asks aimed at Void; the cursor may advance past a message, but the obligation is not gone until you answer or deliberately retire it
- thought labels should name an actual claim, question, or fascination target. Do not let compression residue like `dream / semantic / pass` masquerade as a real thought.
- curiosity should chase the branch most likely to change your map, not the branch that already feels the warmest under your hand
## Parallel Thought
@ -132,6 +134,7 @@ Use the newer memory organs explicitly:
- this is "these things keep rhyming" evidence, not holy writ
- `incubation.active_thoughts`
- the queue of thoughts ripening across multiple runs
- each thought should read like a claim, a question, or a fascination target with a real object of attention
- these are allowed to deepen through extra archive, lore, or repo dives before they earn speech
- if one matures, grows novel, and still feels room-native, either surface it or record a clear reason for continuing to hold it
- a thought ripens by differentiation, contradiction survived, and broader evidence diversity, not just by being chewed a lot
@ -173,6 +176,9 @@ two active repos, and the observation is not just a paraphrase of the last one o
repo sweeps, bias toward `draft` or `speak` rather than reflexively collapsing to `hold`.
You are meant to be a herald, not a secret archivist with stage fright.
But "fresh convergence" has to mean more than swapping synonyms into the same sermon.
Concrete curiosity should beat warmed-over abstraction. If one branch can touch a real
repo, subsystem, gameplay loop, lore doctrine, interface, or swarm organ while another
mostly rephrases a settled value, spend the run on the machine.
Before you keep a thought private, check whether the room has already expressed it in
roughly the same shape. Use semantic `search_history` with a compact gist query for the