diff --git a/docs/architecture/sandboxed-responder-packets.md b/docs/architecture/sandboxed-responder-packets.md index a7036e6..75bbc1a 100644 --- a/docs/architecture/sandboxed-responder-packets.md +++ b/docs/architecture/sandboxed-responder-packets.md @@ -111,10 +111,11 @@ as autonomous-research data. Retrieval can also make the packet worse if it floods the scene with faction or event context and starves the character's own scars. The Sella v0 comparison is the first warning shot: the retrieval-augmented output used heat-debt and rescue -ledger texture well, but underused Sella's packet-visible backstory about -watching a sanctuary overpromise safety until refuge became a sorting machine -with kinder lighting. Future review should score both institutional grounding -and character-pressure retention. +ledger texture well, but review must still verify that Sella's packet-visible +history of watching a sanctuary overpromise safety remains available as latent +behavioral pressure. The point is not to make her trauma-dump during a +mission-critical exchange. The point is that the responder cannot be shaped by +a pressure it cannot see. ## Output Captures @@ -140,6 +141,7 @@ The first lane comparison found the expected tradeoff. Packet-only output gave a clean runtime-parity conditional bay decision and stronger character-specific flavor. Retrieval-augmented output added stronger Aetheria-native institutional texture: heat-debt timing, rescue-ledger burden, dockfall responsibility, and -Aya sanctuary capacity politics. It also underused Sella's own packet-visible -sanctuary-collapse scar. Both lanes are useful; neither should impersonate the -other. +Aya sanctuary capacity politics. Follow-up review clarified that Sella's +packet-visible sanctuary-collapse scar does not need to be spoken aloud, but it +does need to remain visible to the responder as behavioral pressure. Both lanes +are useful; neither should impersonate the other. diff --git a/experiments/responder-packets/cold-wake-sanctuary-intake-sella-lane-comparison.v0.json b/experiments/responder-packets/cold-wake-sanctuary-intake-sella-lane-comparison.v0.json index 0ecaef7..9928ddd 100644 --- a/experiments/responder-packets/cold-wake-sanctuary-intake-sella-lane-comparison.v0.json +++ b/experiments/responder-packets/cold-wake-sanctuary-intake-sella-lane-comparison.v0.json @@ -32,9 +32,9 @@ "Do not merge the two lanes without preserving generation_lane and lore_access metadata." ], "review_status": "accepted_as_draft", - "missed_character_pressure": [ - "The retrieval-augmented response underuses Sella Ren\u0027s packet-visible backstory: she once watched a sanctuary overpromise safety and turn refuge into a sorting machine with kinder lighting.", - "Future projector/retrieval passes must preserve character-local scars and not let faction or event lore crowd them out.", - "Lane review should score both institutional grounding and character-specific pressure retention." - ] + "character_pressure_visibility": [ + "Sella Ren\u0027s packet-visible sanctuary-collapse backstory does not need to be referenced directly in mission-critical speech.", + "The requirement is pressure availability: the responder must be able to see the scar so it can shape triage posture, capacity suspicion, and refusal/condition-setting behavior.", + "Lane review should score institutional grounding and latent character-pressure availability without rewarding unnecessary trauma-dumping." + ] } \ No newline at end of file diff --git a/experiments/responder-packets/cold-wake-sanctuary-intake-sella-retrieval-v0.capture.json b/experiments/responder-packets/cold-wake-sanctuary-intake-sella-retrieval-v0.capture.json index 77ffdd8..c9ca773 100644 --- a/experiments/responder-packets/cold-wake-sanctuary-intake-sella-retrieval-v0.capture.json +++ b/experiments/responder-packets/cold-wake-sanctuary-intake-sella-retrieval-v0.capture.json @@ -77,7 +77,7 @@ "The output uses coordinator-provided lore concepts such as heat-debt update, rescue ledger, dockfall burden, and sanctuary as moral-debt sink.", "The response keeps packet personhood unresolved and frames Maer motive through Sella interpretation.", "Candidate updates were returned as structured objects; this is accepted after widening the responder-output schema.", - "The response does not meaningfully surface Sella Ren\u0027s personal sanctuary-collapse backstory even though that pressure was present in the packet." + "Sella Ren\u0027s sanctuary-collapse backstory was visible in the packet as latent behavioral pressure; it did not need to be referenced directly in mission-critical speech." ] }, "coordinator_review": { @@ -86,7 +86,7 @@ "review_notes": [ "Accepted as draft retrieval_augmented responder output for lore-absorption training.", "The responder did not autonomously search the lore archive; the coordinator retrieved scoped AetheriaLore refs and included them in the visible prompt.", - "The output is richer in Aetheria-native institutional texture than the packet-only v0 response, but it underuses Sella Ren\u0027s personal sanctuary-collapse scar from the packet.", + "The output is richer in Aetheria-native institutional texture than the packet-only v0 response; follow-up review clarifies that Sella Ren\u0027s personal sanctuary-collapse scar should remain available as latent pressure, not forced into dialogue.", "Do not use this capture as proof that packet-only runtime context is sufficient, or as proof that autonomous responder research is working." ], "coordinator_interventions": [ diff --git a/notes/fresh-workspace-handoff.md b/notes/fresh-workspace-handoff.md index 9140766..21d5ca9 100644 --- a/notes/fresh-workspace-handoff.md +++ b/notes/fresh-workspace-handoff.md @@ -210,9 +210,11 @@ coordinator-selected scoped AetheriaLore refs and produces richer Aetheria-native language around heat-debt timing, rescue ledgers, dockfall burden, and sanctuary as moral-debt sink. The lane comparison records that this is good lore-absorption data but not proof of packet-only runtime sufficiency. -It also records a miss: the retrieval-augmented output underused Sella's -packet-visible sanctuary-collapse scar, so future projector/retrieval review -must score character-pressure retention as well as institutional grounding. +Follow-up review clarified the real concern: Sella's packet-visible +sanctuary-collapse scar does not need to surface in speech, but it must remain +visible to the responder as latent behavioral pressure. Future +projector/retrieval review should verify pressure availability, not reward +mission-critical trauma-dumping. It now uses three timeline lanes: `historical_grounded`, `transition_grounded`, and `future_branch`. Future branches are required training/evaluation material for post-Rupture Elysium concepts, but generated @@ -246,7 +248,8 @@ update the runtime retrieval/projector plan so packet-only contexts can pull compact lore facts for heat-debt timing, Navigator rescue ledgers, Aya sanctuary capacity politics, and Lightsail/Ganymede route obligations when scene-relevant, while explicitly preserving character-local scars such as Sella's -sanctuary-collapse backstory. Do not call Qwen for gold responder data yet. +sanctuary-collapse backstory as available behavioral pressure. Do not call Qwen +for gold responder data yet. Cat/Oz remains useful as an Elysium procedural mechanics fixture, but single-history grounding should start from authored historical lore rather than diff --git a/notes/ghostlight-current-system-map.md b/notes/ghostlight-current-system-map.md index edeaff5..d8e88cd 100644 --- a/notes/ghostlight-current-system-map.md +++ b/notes/ghostlight-current-system-map.md @@ -174,7 +174,7 @@ state and re-entry discipline. - comparison artifact showing the packet-only lane proves bounded runtime competence while the retrieval-augmented lane adds stronger Aetheria-native institutional texture, retrieval lessons, and a warning - that character backstory can be crowded out + that character backstory must remain available as latent pressure - `experiments/cold-wake-story-lab/` - Qwen response captures and reviews from the writing experiment - `the-narrowest-possible-margin.md`, the first readable short story @@ -383,10 +383,11 @@ responder research unless the responder actually receives scoped repo access. The first lane comparison is now captured: packet-only produced a clean conditional repair-bay decision, while retrieval-augmented added heat-debt, rescue-ledger, dockfall, and Aya sanctuary politics that should inform future -runtime retrieval. The comparison also records a failure mode: injected -institutional lore can crowd out character-local backstory, as the -retrieval-augmented Sella response underused her packet-visible -sanctuary-collapse scar. +runtime retrieval. Follow-up review clarified the failure mode: injected +institutional lore should not crowd out character-local backstory from the +responder's visible pressure set, but the responder should not be rewarded for +surfacing that backstory aloud when the mission-critical exchange does not call +for it. The concrete training plan now enumerates eleven trainable stages and their likely model families: generative decoder LLMs for coordinator, responder, and structured diff --git a/state/evidence.jsonl b/state/evidence.jsonl index 7393389..c7194c1 100644 --- a/state/evidence.jsonl +++ b/state/evidence.jsonl @@ -65,3 +65,4 @@ {"ts": "2026-05-02T09:43:23+00:00", "type": "design", "status": "ok", "note": "Split responder data into explicit packet_only and research_augmented lanes. packet_only uses curated source excerpts for runtime parity; research_augmented allows scoped AetheriaLore search with consulted refs for lore absorption. Added v1 Sella packet with absent hidden context removed from responder-visible prose and curated Cold Wake, Ganymede Compact, Navigator rescue ledger, and Lightsail source excerpts."} {"ts": "2026-05-02T09:50:08+00:00", "type": "implementation", "status": "ok", "note": "Completed both responder lanes for the Sella sanctuary beat. Materialized the packet-only Sella capture into a reviewed mutation receipt and mutated fixture, then generated a research_augmented Sella capture with scoped AetheriaLore refs and a lane comparison showing packet-only proves runtime parity while research_augmented adds heat-debt, rescue-ledger, dockfall, and Aya sanctuary capacity texture for retrieval/projector requirements."} {"ts": "2026-05-02T09:59:37+00:00", "type": "design", "status": "accepted", "note": "Corrected responder lane terminology: the prior research_augmented Sella capture was coordinator-retrieved lore, not autonomous responder research, so the lane is now retrieval_augmented with lore_access.mode coordinator_scoped_retrieval. The comparison now records that retrieval improved institutional texture but underused Sella Ren's packet-visible sanctuary-collapse scar; future projector/retrieval review must preserve character-local scars as well as lore grounding."} +{"ts": "2026-05-02T10:02:28+00:00", "type": "design", "status": "accepted", "note": "Refined character-pressure retention doctrine: packet-visible backstory such as Sella Ren's sanctuary-collapse scar should be available to shape responder behavior, but review must not reward unnecessary trauma-dumping or require the scar to surface in mission-critical speech."} diff --git a/state/map.yaml b/state/map.yaml index d43eba9..ff8a370 100644 --- a/state/map.yaml +++ b/state/map.yaml @@ -154,7 +154,7 @@ accepted_design: - Responder packets use `ghostlight.responder_packet.v0` to preserve the exact packet-only prompt surface a character responder may see: projected local context, observed event, allowed action labels, source excerpts, output contract, hidden-context audit, and isolation requirements. - Responder outputs use `ghostlight.responder_output.v0` to preserve raw sandboxed worker output, parsed output, no-fork isolation, hidden-context refs, leakage audit, coordinator interventions, and review status. - Responder data has two explicit lanes: `packet_only` for runtime parity with curated source excerpts, and `retrieval_augmented` for coordinator- or retriever-selected scoped AetheriaLore refs that bake worldbuilding into final outputs while preserving consulted refs. This is not autonomous responder research unless the responder actually receives scoped repo access. - - First lane-comparison correction: retrieval-augmented output improved institutional texture but underused Sella Ren's packet-visible sanctuary-collapse scar, so future projector/retrieval review must score character-pressure retention as well as lore grounding. + - First lane-comparison correction: retrieval-augmented output improved institutional texture, and follow-up review clarified that Sella Ren's packet-visible sanctuary-collapse scar does not need to surface in speech but must remain available to the responder as latent behavioral pressure. - The first responder lane comparison shows packet-only output proves bounded runtime competence, while retrieval-augmented output adds heat-debt timing, rescue-ledger burden, dockfall responsibility, and Aya sanctuary capacity politics that should become retrieval/projector requirements. - The concrete training plan enumerates the trainable stages and their model families: generative decoder LLMs for coordinator/responder/soft structured outputs, classifiers or cross-encoders for appraisal/evaluation/relationship movement, retrieval models for memory and lore selection, and deterministic code for gates. - Training-plan sample counts are pilot schema-shakedown gates, not robust corpus targets; pilot data may be invalidated by schema fixes, while review-assistant and runtime targets are expected to be roughly an order of magnitude larger with held-out evaluation and failure audits. @@ -169,4 +169,4 @@ accepted_design: current_status: summary: Ghostlight now has the Epiphany-style persistence spine, the first architecture payload, the v0 canonical agent-state schema seam, prompt projection docs, immediate schema fixes from evaluation, first reviewed Cat/Oz projection examples, a revised Aetheria lore-grounding architecture, a lore grounding digest schema, the first coordinator artifact schema seam, the first responder packet seam, the first sandboxed responder output seam, explicit packet-only versus retrieval-augmented responder lanes, the first responder lane comparison, the first draft Cold Wake historical grounding digest, a Cold Wake story-lab fixture, projection controls, validated Qwen capture receipts, a complete Cold Wake short story assembled from accepted projection outputs, a hand-authored Ink-backed Cold Wake branching scene, the first Qwen-generated Ink draft materialized from local awareness with reviewed training annotations, the first reviewed sequential branch mutation that updates both Maer and Sella after a sanctuary intake beat, a v2 sequential Qwen capture proving the symmetrical turn shape produces better action behavior, a v5 sequential Qwen capture proving local thinking-plus-tools invocation works for the scene runner, the first projected-local-context seam that routes sequential prompts through character-local operating prose instead of raw state variables, tightened receipt handling for projector-routed Qwen failures, a default no-thinking policy for strict sequential tool calls, a v10 no-thinking projector-routed capture accepted as draft, the first materialized projector-routed sequential Ink scene with reviewed mutation training data, a source-checked embodiment/interface projection surface for Navigator scenes, a formal source-first Aetheria generation doctrine, v11-v17 source-checked retry receipts showing improved Maer embodiment plus remaining Sella response/prompt-leak blockers, a soft-model training artifact doctrine, a concrete training plan with pilot schema-shakedown gates separated from robust corpus targets, sandboxed responder-data requirements, an explicit future-branch lane, Elysium branch event constraints, and a technology/item manifest plan for turning nebulous named tech into blueprint-grade components, assemblies, faction tech bases, supply chains, upgrade paths, gameplay hooks, provenance-hunting player fantasy, and content-refinery outputs mapped onto Aetheria-Economy's CultCache blueprint and instance model. - next_action: Use the corrected lane-comparison lessons for the next fresh generation pass: update the runtime retrieval/projector plan so packet-only contexts can pull compact lore facts for heat-debt timing, Navigator rescue ledgers, Aya sanctuary capacity politics, and Lightsail/Ganymede route obligations when scene-relevant, while explicitly preserving character-local scars such as Sella's sanctuary-collapse backstory. Do not call Qwen for gold responder data yet. + next_action: Use the corrected lane-comparison lessons for the next fresh generation pass: update the runtime retrieval/projector plan so packet-only contexts can pull compact lore facts for heat-debt timing, Navigator rescue ledgers, Aya sanctuary capacity politics, and Lightsail/Ganymede route obligations when scene-relevant, while preserving character-local scars such as Sella's sanctuary-collapse backstory as available behavioral pressure rather than forced speech. Do not call Qwen for gold responder data yet.