28 KiB
Ghostlight Current System Map
Ghostlight does not have a runtime yet. The live system right now is the repo's state and re-entry discipline.
Current Control Flow
- A user or agent enters the repo with a bounded question or implementation goal.
- The session rehydrates from:
state/map.yamlnotes/fresh-workspace-handoff.md- this file
notes/ghostlight-implementation-plan.md
- The session runs
npm run state:statusto recover the current summary, next action, and active subgoals. - Work happens in one bounded slice.
- If understanding changes, the map is updated.
- If future belief changes, distilled evidence is appended.
- If compaction pressure rises, the helper audits state hygiene before the session banks the fire.
Live Surfaces
state/map.yaml- canonical experiment, objective, constraints, invariants, active subgoals, accepted design, and current status
state/scratch.md- temporary working memory for one bounded organ
state/evidence.jsonl- durable belief-changing records only
state/branches.json- explicit hypothesis or branch tracking
notes/fresh-workspace-handoff.md- re-entry packet
notes/ghostlight-implementation-plan.md- larger-organ sequence
docs/research/- research backlog and social-model notes
docs/architecture/- world, state, and prompt-system architecture notes
docs/architecture/prompt-projection-contract.md- contract for turning durable state into speaker-local pressure prose and prompt sections
docs/architecture/projection-distillation-plan.md- plan for using a frontier teacher model to generate reviewed projection artifacts before training a smaller student projector
docs/architecture/sequential-agent-runtime.md- scene-local dialogue/runtime loop where agents act from local awareness, generated actions become event proposals, and bounded state mutation drives branchable speech or non-speech scene consequences while preserving training hooks for later open-world, economic, and faction decision models
docs/architecture/ink-branching-scenes.md- integration contract that makes Ink the standard playable branch format while Ghostlight owns local awareness, projection rationale, consequence review, and state mutation around Ink scenes
docs/experiments/cold-wake-story-lab.md- four-scene Cold Wake writing experiment using Ghostlight projection as a story-generation crutch and training-data source
docs/architecture/schema-future-mechanisms.md- parking lot for appraisal, transition, belief, action, conversation, resource, dyad, institution, memory, and embodiment mechanisms not yet promoted into the v0 schema
docs/architecture/aetheria-lore-grounding-architecture.md- boundary and plan for grounding single-history training data in authored
Aetheria history while treating Elysium gameplay-era outcomes as
conditional branch canon;
includes the source-first doctrine for checking AetheriaLore and patching
blocking lore gaps before generation, plus a
future_branchlane for reviewed possible post-Rupture Elysium futures
- boundary and plan for grounding single-history training data in authored
Aetheria history while treating Elysium gameplay-era outcomes as
conditional branch canon;
includes the source-first doctrine for checking AetheriaLore and patching
blocking lore gaps before generation, plus a
docs/architecture/lore-grounding-digest-format.md- human-facing format guide for cultural, factional, institutional, role, and speaker-boundary lore digests
docs/architecture/qwen-invocation-notes.md- verified local Qwen invocation policy: Ollama
/api/chat, native tools, thinking enabled, repair for known nested-argument double-stringify cases
- verified local Qwen invocation policy: Ollama
docs/architecture/projected-local-context.md- contract for the projector seam that turns canonical state into character-local operating context before a response model sees it, including source-checked embodiment and interface affordances, runtime retrieval requirements, and latent pressure handling
docs/architecture/soft-model-training-artifacts.md- doctrine for treating every non-deterministic judgment as reviewed training material for future specialized models, including coordinator/story runtime, projector, character agent/responder, appraiser, state mutator, relationship/perception classifier, Aetheria responder, and institution/faction/consumer decision models
docs/architecture/sandboxed-responder-packets.md- contract for the exact packet-only surface handed from an omniscient
coordinator to a sandboxed character responder, including visible event,
local prompt, allowed actions, source excerpts, hidden-context audit, and
output contract; also defines the separate
packet_onlyandretrieval_augmentedresponder-data lanes
- contract for the exact packet-only surface handed from an omniscient
coordinator to a sandboxed character responder, including visible event,
local prompt, allowed actions, source excerpts, hidden-context audit, and
output contract; also defines the separate
docs/architecture/training-plan.md- concrete enumeration of trainable Ghostlight stages, including each stage's inputs, outputs, likely model architecture, artifact family, first training gate, and timeline coverage across historical grounded, transition grounded, and future-branch examples
docs/architecture/technology-item-manifest-plan.md- plan for turning coordinator-led Aetheria exploration into game-usable
item, component, assembly, faction tech-base, prerequisite, supply-chain,
and upgrade-path data, including elaboration of existing nebulous named
technologies into blueprint candidates; now explicitly targets the
Aetheria-Economy CultCache
ItemDatatechnological blueprint classes instead of a parallel Ghostlight item database
- plan for turning coordinator-led Aetheria exploration into game-usable
item, component, assembly, faction tech-base, prerequisite, supply-chain,
and upgrade-path data, including elaboration of existing nebulous named
technologies into blueprint candidates; now explicitly targets the
Aetheria-Economy CultCache
docs/architecture/aetheria-cold-wake-training-fixture.md- human-facing note that reframes the recovered Cold Wake scenario as historical projection feedstock rather than the active product target
schemas/agent-state.schema.json- v0 canonical agent-state contract for agents, relationships, events, scenes, perceived overlays, and dialogue context packs
schemas/projection-example.schema.json- v0 reviewed projection artifact contract for speaker-local prompt projection examples, including projection controls for frame, authority, object custody, required semantics, and forbidden resolutions
schemas/lore-grounding-digest.schema.json- v0 contract for grounding historical fixtures in source-backed cultural and factional pressure fields
schemas/projected-local-context.schema.json- v0 contract for projected character-local operating context artifacts, including structured runtime retrieval requirements and latent pressure requirements
schemas/coordinator-artifact.schema.json- v0 contract for coordinator/story-runtime receipts: scene setup, next-beat choice, acting-agent choice, machinery invocations, sandboxed responder handoffs, world-state refs, proposed deltas, branch constraints, item-manifest deltas, unresolved hooks, glue prose, and review notes
schemas/responder-packet.schema.json- v0 contract for the exact input a sandboxed character responder may see,
now lane-labeled as
packet_onlyorretrieval_augmentedwith explicit lore-access policy; research-enabled packets can require responder-scoped repository search with visible research instructions
- v0 contract for the exact input a sandboxed character responder may see,
now lane-labeled as
schemas/responder-output.schema.json- v0 contract for reviewed captures from sandboxed responder workers, preserving raw output, parsed output, isolation method, leakage audit, and coordinator review; output captures also preserve generation lane and consulted lore refs, and research-enabled captures can preserve a research summary
schemas/agent-state.required-fields.json- required first-class variable names for canonical state and relationship stance
examples/agent-state.call-of-the-void.json- first Call of the Void-flavored fixture for schema and projection work
examples/agent-state.cold-wake-story-lab.json- first story-shaped historical fixture for Cold Wake projection work
examples/projections/call-of-the-void.scene-broken-taxi-oz.jsonl- first reviewed Cat and Oz projection examples
examples/projections/call-of-the-void.scene-broken-taxi-oz.pretty.json- pretty-printed mirror for inspecting the first projection examples
examples/projections/cold-wake-story-lab.jsonl- response-turn projection examples for the Cold Wake story lab, including rejected and accepted Maer/Veyr prompt revisions
examples/coordinator/cold-wake-sanctuary-intake.v0.json- first coordinator artifact schema-shakedown example, backfilled from the accepted-as-draft sanctuary packet-assessment beat and explicitly marked as not gold responder data
examples/responder-packets/scene-02-sanctuary-intake.sella_ren.packet.v0.json- first responder-packet schema-shakedown example for Sella's sanctuary response; accepted as draft for sandbox handoff testing, not gold responder output
examples/responder-packets/scene-02-sanctuary-intake.sella_ren.packet.v1.json- corrected packet-only Sella handoff with absent hidden context moved out of responder-visible prose and curated AetheriaLore excerpts added for Cold Wake, Ganymede rescue obligations, Navigator rescue ledgers, and Lightsail reliability
examples/responder-packets/scene-02-sanctuary-intake.sella_ren.packet.research.v0.json- first concrete research-enabled Sella handoff; it uses
responder_scoped_repository_search, declares the allowed AetheriaLore scope, permits repo access only for that scope, and requires consulted refs plus research summary in the responder object
- first concrete research-enabled Sella handoff; it uses
experiments/responder-packets/cold-wake-sanctuary-intake-sella-v0.capture.json- first no-fork sandboxed responder output capture from the Sella packet; accepted as draft with no leakage flags and no coordinator prose repair
experiments/responder-packets/cold-wake-sanctuary-intake-sella-v0.mutation.json- reviewed mutation receipt that materializes the packet-only Sella response into scene-local state, relationship, memory, perceived-overlay, and unresolved-hook updates without resolving packet personhood
experiments/responder-packets/cold-wake-sanctuary-intake-sella-retrieval-v0.capture.json- first retrieval-augmented responder capture using coordinator-selected scoped AetheriaLore refs for Aya sanctuary politics, Cold Wake heat-debt pressure, Navigator rescue ledgers, Ganymede route obligations, and Lightsail reliability
experiments/responder-packets/cold-wake-sanctuary-intake-sella-lane-comparison.v0.json- 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 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 assembled from accepted projection outputs
examples/lore-grounding/historical-flashpoint.template.json- fillable template for the first historical Aetheria grounding digest
examples/lore-grounding/cold-wake-panic.ganymede-corridor.json- first draft historical grounding digest, centered on Cold Wake Panic in the Ganymede/Jovian corridor
examples/ink/cold-wake-sanctuary-intake.ink- first playable Ink-backed Cold Wake branching scene, using speech, gesture, object revelation, and moral-pressure branches
examples/ink/cold-wake-sanctuary-intake.training.json- reviewed sidecar annotation tying Ink branches to Ghostlight local awareness, projection controls, branch rationale, consequence surfaces, and manual mutation policy
examples/ink/cold-wake-sanctuary-intake.qwen-draft.ink- first Qwen-generated playable Ink draft materialized from a Ghostlight local-awareness prompt
examples/ink/cold-wake-sanctuary-intake.qwen-draft.training.json- sidecar annotation for the generated draft, accepted as training material rather than final polished prose
examples/ink/cold-wake-sanctuary-intake.qwen-sequential-v10.ink- first playable Ink branch materialized from an accepted projector-routed sequential Qwen capture
examples/ink/cold-wake-sanctuary-intake.qwen-sequential-v10.training.json- sidecar annotation for the v10 materialized sequential branch
experiments/ink/cold-wake-sanctuary-intake-qwen-branch-candidates-v1.capture.json- reviewed Qwen branch-generation receipt, including prompt, parsed response, strengths, failure notes, and pipeline lessons
experiments/ink/cold-wake-sanctuary-intake-qwen-sequential-v1.capture.json- useful-needs-revision sequential Qwen receipt that separates Maer-local choice generation from Sella-local next-action generation, while preserving the discovered failure around missing per-turn appraisal/consolidation and invalid action labels
experiments/ink/cold-wake-sanctuary-intake-qwen-sequential-v2.capture.json- useful-needs-revision sequential Qwen receipt using the symmetrical turn
model; it produced canonical action labels and a concrete Sella
withhold_objectnext action, but still needs stricter schema enforcement because some array fields came back as strings or objects
- useful-needs-revision sequential Qwen receipt using the symmetrical turn
model; it produced canonical action labels and a concrete Sella
experiments/ink/cold-wake-sanctuary-intake-qwen-sequential-v3.capture.json- useful-needs-revision thinking-plus-tools receipt that proved Maer could call the tool while later passes still fell back to plain JSON content because old prompt text conflicted with the tool path
experiments/ink/cold-wake-sanctuary-intake-qwen-sequential-v5.capture.json- accepted-as-draft thinking-plus-tools receipt using Ollama
/api/chat, native tools, andthink: truefor Maer choice, Sella appraisal, and Sella next action
- accepted-as-draft thinking-plus-tools receipt using Ollama
experiments/ink/cold-wake-sanctuary-intake-qwen-sequential-v6-projector.capture.json- useful-needs-revision projector-routed receipt proving Qwen prompts can use
character-local operating prose instead of raw state variables; still
exposed nested
response_constraintsstringification in appraisal output
- useful-needs-revision projector-routed receipt proving Qwen prompts can use
character-local operating prose instead of raw state variables; still
exposed nested
experiments/ink/cold-wake-sanctuary-intake-qwen-sequential-v8-projector-tightened.capture.json- useful-needs-revision receipt proving failed Maer choice generation is now
captured instead of crashing the runner; exposed malformed stringified
choices
- useful-needs-revision receipt proving failed Maer choice generation is now
captured instead of crashing the runner; exposed malformed stringified
experiments/ink/cold-wake-sanctuary-intake-qwen-sequential-v9-projector-tightened.capture.json- useful-needs-revision receipt showing Qwen can still drop tool calls entirely under projector-routed prompting
experiments/ink/cold-wake-sanctuary-intake-qwen-sequential-v10-no-think.capture.json- accepted-as-draft receipt showing projector-routed sequential generation works cleanly when strict tool calls run with thinking disabled
examples/projected-contexts/- current rendered projected local context artifacts for Maer and Sella in the sanctuary intake scene
experiments/ink/qwen-chat-tools-smoke.json- smoke receipt proving local
qwen3.5:9breturns bothmessage.thinkingand nativetool_calls
- smoke receipt proving local
experiments/ink/cold-wake-sanctuary-intake-qwen-sequential-v1.mutation.json- first reviewed branch mutation receipt, including manual participant appraisal/consolidation, normalized action labels, applied state deltas, and deferred unresolved facts
examples/agent-state.cold-wake-story-lab.after-sanctuary-ledger.json- mutated Cold Wake fixture after the selected sanctuary ledger branch, updating Maer and Sella state, relationship stance, memories, perceived overlays, event log, and active scene memories
examples/agent-state.cold-wake-story-lab.after-v10-packet-assessment.json- mutated Cold Wake fixture after the v10 packet-assessment branch, preserving unresolved packet personhood while updating state, relationship stance, memories, perceived overlays, event log, and active scene memories
experiments/ink/cold-wake-sanctuary-intake-qwen-sequential-v10.mutation.json- reviewed mutation receipt for v10 materialization, keeping fuzzy appraisal and relationship movement manual and auditable
tools/ghostlight_state.py- compact state inspection and evidence or branch updates
tools/ghostlight_prepare_compaction.py- pre-compaction audit of required files, handoff content, and git hygiene
tools/validate_agent_state.py- dependency-free validator for the current schema fixture invariants
tools/validate_projection_examples.py- dependency-free validator for projection example JSONL records
tools/run_qwen_projection.py- sends a projection prompt to the LAN Qwen box and saves response/capture artifacts without PowerShell encoding footguns
tools/validate_qwen_captures.py- validates reviewed Qwen response capture receipts
tools/validate_lore_grounding.py- dependency-free validator for lore grounding digest examples
tools/validate_ink_examples.py- compiles Ink examples with
inkjsand validates sidecar training annotations plus reviewed Qwen Ink-generation captures
- compiles Ink examples with
tools/run_qwen_ink_branch_generation.py- builds a local-awareness prompt from the Cold Wake fixture and asks Qwen for Ink branch candidates
tools/materialize_qwen_ink_draft.py- turns a reviewed Qwen branch capture into a playable Ink draft and sidecar annotation
tools/materialize_sequential_capture.py- turns an accepted sequential capture into playable Ink, sidecar annotation, reviewed mutation receipt, and a mutated fixture
tools/run_qwen_ink_sequential_generation.py- builds a sequential Qwen prototype through Ollama
/api/chatwith native tools and thinking disabled by default: actor-local choices, participant appraisal/consolidation, then next-actor action from updated state; now routes character context through the projected-local-context seam, captures failed Maer choice calls, separates repair notes from fatal validation notes, and keeps numeric appraisal deltas out of next-action prompt context
- builds a sequential Qwen prototype through Ollama
tools/check_qwen_chat_tools.py- fast smoke test for local Qwen thinking plus native tool calling
tools/project_local_context.py- deterministic first projector that compiles canonical state, scene facts, relationships, memories, projection controls, and optional observed event into character-local operating prose; for cetacean Navigators it projects source-checked habitat affordances from AetheriaLore instead of generic human body defaults; now emits scene-triggered compact lore retrieval requirements and latent character-history pressure requirements
tools/validate_projected_contexts.py- validates projected context artifacts and rejects prompt text that leaks raw canonical state internals
tools/validate_coordinator_artifacts.py- validates v0 coordinator artifacts, including sandboxed responder handoff fields, world-state refs, branch constraints, proposed deltas, unresolved hooks, and review status
tools/build_responder_packet.py- builds packet-only or research-enabled responder handoffs from a coordinator artifact and projected local context; source excerpts now come from active projected runtime retrieval requirements instead of a static lore list
tools/validate_responder_packets.py- validates responder packets and rejects prompt surfaces that leak raw state internals or hidden coordinator context markers
tools/validate_responder_outputs.py- validates sandboxed responder output captures, including exact raw-to-parsed JSON agreement, no-fork isolation, action labels, leakage audit, and review metadata; also validates responder mutation receipts
tools/apply_responder_output_mutation.py- applies a reviewed responder-output capture into a mutated agent-state fixture and mutation receipt
tools/apply_sequential_ink_branch_mutation.py- applies a reviewed replay of one selected branch into a mutated agent-state fixture and mutation receipt without letting Ink variables become canonical truth
tools/build_cold_wake_story_lab_fixture.py- generator for the verbose Cold Wake story-lab state fixture
What Does Not Exist Yet
- a classifier pipeline
- a prompt renderer
- a full projection training corpus
- a deterministic projection input slicer
- a student projection model
- a simulation/event loop implementation
- a relationship engine
- a culture prior engine
- an automatic Ink branch generator from local awareness
- automatic promotion of Ink branch outcomes into canonical state
The canonical agent-state schema, projection example schema, lore grounding
digest schema, projected-local-context schema, coordinator artifact schema,
responder-packet schema, first draft Cold Wake grounding digest, and first Cold Wake story-lab Qwen
captures now exist as v0 seams. The
first complete Cold Wake story pass exists with receipts, projection controls
have been promoted, and Ink is now the standard branching scene format. The
first Qwen-generated Ink draft now exists with a reviewed capture, the first
selected branch now has a reviewed mutation replay that updates both involved
characters, v2 shows the symmetrical turn model improves action behavior, v5
proves local thinking-plus-tools invocation works for the scene runner, and v6
routes that runner through projected character-local operating context. v8 and
v9 show that Qwen tool-call reliability, not the projector seam, is now the
immediate blocker. The v9 thinking trace also showed schema self-check looping
instead of tool-call completion, so strict sequential generation now disables
thinking by default. v10 validated that correction with an accepted-as-draft
projector-routed no-thinking capture, which has now been materialized into Ink
plus reviewed mutation training data. The next implementation target is
fixing the Sella next-action prompt/rendering path after source-checked v11-v17
retries improved Maer's Navigator embodiment but exposed object-custody drift,
invalid appraisal paths, Qwen tool dropout, and prompt-constraint leakage into
responder prose. The soft-model artifact doctrine now makes the broader
training boundary explicit: deterministic gates stay code, while fuzzy
coordination, projection, action choice, appraisal, mutation, relationship
movement, Aetheria responder training, and economic/faction decision judgments
become reviewed artifacts for later fine-tuning or distillation. Gold responder
data now requires a sandbox boundary: the coordinator may be omniscient, but the
responder sees only the exact projected local packet, visible event, allowed
actions, and explicitly included source excerpts; artifacts preserve raw output,
reviewed output, hidden-context refs, leakage audit, isolation method, and
coordinator intervention labels. The responder packet seam now turns that
requirement into a concrete artifact and validator, so the next gold-data pass
can hand a worker the exact prompt surface instead of hoping nobody leaks the
coordinator's brain into the room. The first no-fork Sella responder capture now
exists and validates as accepted-as-draft with raw output preserved, parsed
output matched exactly, leakage audit clear, and no coordinator prose repair.
Responder data now 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 can bake setting priors into
final responses while preserving consulted refs. This is not autonomous
responder research unless the responder actually receives scoped repo access.
That explicit repo-access path is now named
responder_scoped_repository_search: the packet must tell the responder to
consult the declared lore docs before answering, show the allowed scope, and
give behavior-grounding research instructions. The capture must record consulted
refs and a research summary.
The first concrete research-enabled Sella packet has now been generated and
validates alongside the packet-only handoffs.
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. 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 runtime retrieval/projector correction is now concrete: projected contexts
carry source-backed retrieval requirements and latent pressure requirements,
and the responder packet builder copies active retrieval requirements into
packet source excerpts. The first Sella packet now receives Cold Wake heat-debt,
Navigator rescue-ledger, Aya sanctuary-capacity, and Ganymede/Lightsail route
obligation facts through that seam.
The concrete training plan now enumerates
eleven trainable stages and their likely model
families: generative decoder LLMs for coordinator, responder, and structured
soft outputs; classifiers or cross-encoders for appraisal, evaluation, and
relationship movement; embedding/retrieval models for memory and lore selection;
and deterministic code for gates. The listed per-stage counts are pilot
schema-shakedown gates, not robust training targets; review-assistant and runtime
targets are larger corpus tiers meant to avoid declaring victory after the first
small pile of sacrificial data merely shows where the schema is wrong. The
first coordinator artifact seam now validates through npm run schema:validate
and backfills the sanctuary packet-assessment beat as an accepted-as-draft
schema shakedown example. The coordinator is currently the authoring
agent: it glues scenes together, carries world-state refs and unresolved hooks,
decides which machinery runs, and emits glue prose. That output should stay
game-engine-shaped even while the prose lives in model imagination. The
lore/tone goal is not a separate adapter; it is a responder trained on strong
timeline-spread Aetheria responses until the setting's assumptions become
native priors. The timeline plan now includes generated possible futures in
Elysium as reviewed future_branch artifacts so models learn post-Rupture
concepts as branch-local canon indexed by lineage. It also distinguishes
branch attractors, fated events, technology-order constraints, quest injections,
and ordinary branch-local events so Elysium can support likely faction births,
authored rails, discovery order, and quest injection without flattening the
timeline into equal-probability soup. The technology/item manifest plan now
requires exploration to emit item families, variants, assemblies,
subassemblies, components, materials, faction tech-base access, prerequisites,
bottlenecks, compatibility rules, supply-chain consequences, and quest hooks for
both pre-Elysium starting tech and post-Rupture innovations, mapped onto
Aetheria-Economy's existing CultCache technological blueprint classes:
SimpleCommodityData, CompoundCommodityData, ConsumableItemData, GearData,
and other EquippableItemData subclasses. Runtime ItemInstance classes remain
simulation, inventory, cargo, provenance, branding, and world-state objects.
Recipes, assembly trees, compatibility rules, tooling, facilities, process
requirements, and supply-chain dependencies are blueprint metadata or
engine-schema gaps attached to the relevant *Data class until the economy code
has explicit storage for them. The player-facing gear fantasy is provenance
hunting: players chasing an ultimate item should trace exceptional assemblies,
manufacturer processes, repair history, counterfeit substitutions, supply-chain
lineage, and operating constraints through the galaxy. The technology pipeline is
therefore a content refinery: it grounds ideas in lore, decomposes them into
parts and supply chains, maps them to economy blueprint classes, defines
performance and instance variation, preserves review labels, and emits
database-shaped candidates instead of spreadsheet sludge. Do not materialize
another capture until it has no failure or repair notes.