Treat conditional Face speech as silence

notes/fresh-workspace-handoff.md

@@ -154,6 +154,7 @@ Git history and smoke artifacts carry routine proof. `state/evidence.jsonl` shou
 - Later on 2026-05-22, Mimir started looping on an already-acknowledged correction. Root cause: the Face prompt did include recent Mimir repo activity, but nearby-channel sections obscured cross-channel order and Mimir's typed state still carried active correction-pressure, so the Face and Interpreter treated a resolved repair as live debt. The repair added a deterministic visible cross-channel chronology to repo Face prompts, made the Interpreter treat later same-Face acknowledgements as resolved, preserved manual `blocked` status during participant reconciliation, and retired Mimir's duplicated correction-pressure state plus the stale v1 OBS/SRT drift-ledger incubation. The isolated Face/Interpreter smoke produced no SAY for the repeated acknowledgement.
 - Immediately after, Heimdall produced an Aquarium-shaped reply that the parent Interpreter normalized into `channel: development`, making the post look contextless. Root cause: the Interpreter was allowed to translate a natural `Would say` into both content and venue, and worker fallback treated the scheduled output channel as the default mouth. The repair makes conversation channel ownership deterministic: `SAY` defaults to `current_room`, `current_room` resolves to `job.guildContext.channelId`, owner-DM aliases fall back to the conversation channel unless explicitly owner-private, and the Interpreter prompt says not to invent a speech venue merely because the topic is domain-shaped. The stack was rebuilt and restarted from `dist` after typecheck/build passed.
 - Huginn/Libby then showed a separate echo failure: Huginn mirrored Heimdall's bridge line in schema language, and Libby mirrored Huginn's schema line back into Aquarium. Root cause is not channel routing; it is the parent Interpreter accepting agreement as speech when it adds no new question, source/repo fact, disagreement, handoff, or social move. The Interpreter now has an echo gate and must retry/drop sibling-shaped public speech; Huginn's echo-born Libby bond, bridge/status read, and deleted `#development` delivery receipt were retired via typed Face-state operations in `CultCacheTS`.
+- A later Huginn turn exposed a second Interpreter routing bug: the Face wrote `Would say: Nothing in aquarium unless...`, which was a conditional explanation of silence, but the parent Interpreter treated the presence of `Would say` as an unconditional public message. The prompt now says conditional/negative `Would say` lines route without `SAY` while preserving any durable pressure as a STATE NOTE.
 - Do not restart the swarm by momentum in future sessions; it is already running. If it must be stopped again, use `state/agent-swarm-paused.json` and the orchestrator task deliberately, then update this handoff.
 - Direct Face `UPDATE REQUEST` and `BIFROST TOPIC` packets are ignored while `REPO_FACE_BIFROST_ENABLED=false`. Do not debug missing Bifrost jobs as a live failure unless that flag has been deliberately enabled again.
 - Live model-facing prompt prose has started moving out of executable code. The shared loader is `packages/shared/src/prompt-template.ts`; extracted prompt surfaces live under `prompts/`, including repo Face turn/reviewer prompts, provider wrapper prompts, situational social read prompts, source-grounding reminders, consensus-feed prompts, and repo Face model-smoke scenarios. Code should project data and select prompt names; prompt files should own behavioral prose.

prompts/repo-face-turn-interpreter.prompt.md

@@ -43,6 +43,7 @@ Critical examples:
 - Use retry when a public SAY responds to a correction by shrinking the corrected demand instead of actually changing shape. Example: if someone says manual review receipts are wrong for an automated ML loss loop, a smaller "one listening canary" is still a relapse, not a valid acknowledgement.
 - Use retry when a public SAY invents culture/doctrine around a repeated phrase that the transcript shows is still contested or freshly corrected. The Face may argue for a value in-character, but it may not present the value as room law or shared religion without uptake.
 - Use drop when a second attempt is still bad, unsafe, empty, or not worth routing.
+- Use route-without-SAY, not retry, when the Face explicitly holds public speech back: "nothing in this room", "nothing public", "stay private", "not unless", "only if", "when X happens", or any equivalent conditional/negative speech intent. Preserve any durable pressure as STATE NOTE if useful, but do not turn the explanation of silence into the public message.
 - Do not emit governance or dispatch blocks. If the Face wants work done and consensus or direct approval is already clear, save that as STATE NOTE and let the room-facing SAY mention the concrete next step plainly. ARTICLE is allowed only for a complete bylined draft as described above.
 
 Attempt: {{attempt}}
@@ -83,7 +84,9 @@ Normalized output rules:
 - Use STATE NOTE kind `status` when the Face reads their standing, another person's standing, attention politics, consultation/bypass, pampering, neglect, challenge, admiration, or threat.
 - If public speech is warranted, emit one SAY block.
 - If a complete bylined article draft is warranted, emit one ARTICLE block. The worker owns YAML frontmatter rendering; do not put markdown frontmatter in the body.
-- If the Face output includes a `Would say` line, a direct answer to a live room invitation, or a clear desire to respond to a human asking the agents to speak, route that as SAY unless it is unsafe or empty.
+- A `Would say` line is only public speech when it contains an unconditional message intended for now.
+- Do not emit SAY for a `Would say` line that says the Face would say nothing, hold silence, wait, speak only if/unless/when a future condition happens, or otherwise withhold public speech. Example: `Would say: Nothing in aquarium unless the cleanup produces a specimen` routes no SAY; it may become a STATE NOTE about specimen pressure.
+- If the Face output includes an unconditional `Would say` line, a direct answer to a live room invitation, or a clear desire to respond to a human asking the agents to speak, route that as SAY unless it is unsafe or empty.
 - If the Face visibly acknowledges a human correction of its own prior claim/proposal, route that acknowledgement as SAY unless unsafe, duplicative, or contradicted by the rest of the Face output. Acknowledgement is social repair only when the repaired claim actually changes.
 - Route correction acknowledgements to the room where the correction happened, normally the `Current room (...)` named in the Face prompt. Do not move an acknowledgement to a domain/work channel merely because the Face also mentioned a future work/proposal there. If you are choosing between `aquarium` and `development` for an acknowledgement, choose `aquarium` unless the human correction itself happened in `development`.
 - Do not invent a speech venue. If the Face output gives a `Would say` or natural public reply without explicitly naming a channel, emit `channel: current_room`. Topic relevance is not permission to move the reply into a domain channel; a context-shaped reply posted elsewhere becomes orphaned noise.

state/evidence.jsonl

@@ -294,3 +294,4 @@
 {"ts":"2026-05-22T22:09:00Z","type":"maintenance","status":"ok","note":"Retired Huginn's deleted/contextless #development delivery receipt in CultCacheTS after the echo-gate fix so the active recent-speech surface no longer treats the removed post as live context."}
 {"ts":"2026-05-22T22:22:23Z","type":"fix","status":"ok","note":"Repo Face CTB idle cooling is scheduler-owned: watched Discord human inactivity now clamps unprompted swarm dispatches to one per idle recovery window while pending mentions bypass the brake. Dry-run on 2026-05-22 reported recent human activity, idleCooling.active=false, and next unprompted dispatch gate in output."}
 {"ts":"2026-05-22T22:37:53Z","type":"fix","status":"ok","note":"Nibu misgendered Metacrat because explicit she/her profile guidance was only fed into the repo Face state projector, and the projector omitted it from the final Face prompt. Repo Face prompts now render deterministic human pronoun guidance outside the projected memory surface, twice for salience, and pronoun evidence display prefers explicit/resolved-supporting evidence over contradictory ambient reads. Retired the bad Nibu delivery receipt from AetheriaLore typed state."}
+{"ts":"2026-05-22T22:45:09Z","type":"fix","status":"ok","note":"Huginn's confusing Aquarium post came from the parent Interpreter treating a conditional silence line as speech: the Face wrote 'Would say: Nothing in aquarium unless...' while asking for a CultCacheTS specimen boundary. Interpreter prompt now routes conditional/negative Would say lines without SAY and preserves any durable pressure as STATE NOTE instead."}

state/map.yaml

@@ -149,6 +149,7 @@ accepted_design:
   - Repo Face introductions are a first-public-speech obligation, not a second-turn reward. If a Face has no public speech receipt or durable memory that it introduced itself, its next warranted public post should introduce the Face naturally in character, even when `queuedCount` is zero.
   - Public repo Face speech must not expose scheduler/provenance machinery. Labels such as `Repo-face heartbeat from ...`, `heartbeat complete`, or repo-name diagnostic prefixes are prompt smell and worker-posting smell; webhook persona already supplies identity.
   - Public repo Face speech must remain in the conversation that gave it meaning unless the Face explicitly chose another registered channel. The parent Interpreter translates natural `Would say` intent into action DSL, but it does not own venue selection; if the child Face did not name a channel, the worker clamps `SAY` to `current_room`, and `current_room` resolves to the job's `guildContext.channelId` before channel allow-list checks. Domain relevance is not permission to move an Aquarium-shaped reply into `#development`.
+  - Parent Interpreter speech routing must distinguish a Face's public intent from its explanation of silence. `Would say` is not automatically a side effect: conditional or negative lines such as "nothing in Aquarium unless X" route without `SAY`, while durable pressure may become a typed state note. Otherwise private restraint gets laundered into public noise.
   - Agreement between Faces must earn its airtime. The parent Interpreter should retry or drop a public `SAY` that merely restates a recent agent claim in another Face's vocabulary without adding a concrete question, source/repo fact, disagreement, handoff, or social move. Echo-formed bonds/status reads are state poison and should not be preserved.
   - Void has explicit herald jurisdiction over the GameCult website and blog, especially `gamecult-site`. Routine rumination still must not edit tracked files directly, but essay-shaped pressure should queue an article-proposal candidate naming the title/angle, target repo/topic, and whether the piece is consensus-seeking or Void-bylined perspective.
   - Repo-activity cursor advancement is parent-owned by successful normal rumination passes. The child sees read-only repo weather, but the runner records typed repo cursor updates afterward so old commits do not keep reappearing as fresh weather.