Collapse memory graph read adapter

This commit is contained in:
Meta Vens 2026-05-18 05:57:04 +01:00
commit 77c62fb488
4 changed files with 41 additions and 71 deletions

View file

@ -39,7 +39,7 @@ remembers doctrine; the branch remembers the blade.
- `epiphany-openai-runtime` worker result ingress now parses assistant output directly into typed role/reorient ingress structs selected from the durable launch document, stores nested `statePatch` and `selfPatch` as typed MessagePack in `EpiphanyRuntimeRoleWorkerResult`, and no longer passes worker result data through generic `serde_json::Value`. JSON survives there only as the OpenAI text-output edge before typed deserialization.
- Agent self-memory review/application now has typed `AgentSelfPatch` core APIs. JSON-value selfPatch functions are compatibility wrappers for CLI/artifact ingress; native callers should parse once at the edge and then call `review_agent_self_patch_document` / `apply_agent_self_patch_document`. Role-result self-persistence no longer exports a JSON-value reviewer; the live API is `review_role_self_patch_document`.
- `epiphany-openai-codex-spine` no longer uses Codex's Responses client, model-provider layer, `codex-api`, `codex-model-provider`, `codex-model-provider-info`, or `codex-protocol` for outbound model requests. It reaches `codex-login` only through `epiphany-openai-auth-spine` so credential loading/refresh stays Codex-compatible while request/status/event/runtime surfaces stay typed Epiphany documents.
- `notes/epiphany-memory-graph-unified-plan.md` is the current memory architecture direction drawn from `Fractal Domains And The Cache That Bites` plus the VoidBot memory scar tissue. The repo semantic architecture/dataflow graph and agent memory are one `EpiphanyMemoryGraph` substrate with repo, role, short-term, incubation, agency, candidate-intervention, and evidence profiles. The first native skeleton is landed: `epiphany-state-model` owns shared graph document types, `epiphany-core::memory_graph` owns stable ids, validation, freshness, context cuts, and CultCache persistence, and `epiphany-memory-graph` can status/validate/context/semantic-context/smoke/index the store. `epiphany-core::semantic_cache` now owns the shared Qdrant/Ollama transport used by workspace retrieval and memory graph indexing, so Qdrant remains a rebuildable contribution cache for typed graph documents, not truth. Runtime role results expose typed `memoryPatchCandidates`; Modeling `roleAccept` applies them to `state/memory-graph.msgpack`; native role/reorientation launches now prefer that persisted memory graph for bounded typed `memoryContext` cuts and only fall back to legacy accepted thread graphs for bootstrap. Reorientation launch no longer carries full `graphs` cargo. The old `thread/epiphany/propose` read route no longer returns replacement `graphs`; it exposes graph growth as typed `memoryPatchCandidates` and withholds frontier updates that depend on unapplied memory growth. Generic `thread/epiphany/update` and `thread/epiphany/promote` now reject `patch.graphs` at the bridge. `thread/epiphany/view`, `thread/epiphany/freshness`, `thread/epiphany/context`, and `thread/epiphany/graphQuery` now prefer the persisted memory graph for loaded threads and project back into legacy graph-shaped response lenses only at the bridge edge. `notes/repo-fractal-dataflow-cache-plan.md` and `notes/agent-memory-fractal-cache-plan.md` are profile plans, not rival organs.
- `notes/epiphany-memory-graph-unified-plan.md` is the current memory architecture direction drawn from `Fractal Domains And The Cache That Bites` plus the VoidBot memory scar tissue. The repo semantic architecture/dataflow graph and agent memory are one `EpiphanyMemoryGraph` substrate with repo, role, short-term, incubation, agency, candidate-intervention, and evidence profiles. The first native skeleton is landed: `epiphany-state-model` owns shared graph document types, `epiphany-core::memory_graph` owns stable ids, validation, freshness, context cuts, and CultCache persistence, and `epiphany-memory-graph` can status/validate/context/semantic-context/smoke/index the store. `epiphany-core::semantic_cache` now owns the shared Qdrant/Ollama transport used by workspace retrieval and memory graph indexing, so Qdrant remains a rebuildable contribution cache for typed graph documents, not truth. Runtime role results expose typed `memoryPatchCandidates`; Modeling `roleAccept` applies them to `state/memory-graph.msgpack`; native role/reorientation launches now prefer that persisted memory graph for bounded typed `memoryContext` cuts and only fall back to legacy accepted thread graphs for bootstrap. Reorientation launch no longer carries full `graphs` cargo. The old `thread/epiphany/propose` read route no longer returns replacement `graphs`; it exposes graph growth as typed `memoryPatchCandidates` and withholds frontier updates that depend on unapplied memory growth. Generic `thread/epiphany/update` and `thread/epiphany/promote` now reject `patch.graphs` at the bridge. `thread/epiphany/view`, `thread/epiphany/freshness`, `thread/epiphany/context`, and `thread/epiphany/graphQuery` now prefer the persisted memory graph for loaded threads through one read-route store adapter and project back into legacy graph-shaped response lenses only at the bridge edge. `notes/repo-fractal-dataflow-cache-plan.md` and `notes/agent-memory-fractal-cache-plan.md` are profile plans, not rival organs.
- Durable Epiphany state types now live in root crate `epiphany-state-model`. `codex-protocol` re-exports them only so old rollout/API callers keep compiling; do not add new durable `Epiphany*` state structs under `vendor/codex`. `epiphany-core` imports the native model directly and no longer depends on `codex-protocol`. Pressure derives from native `EpiphanyTokenUsageSnapshot`; Codex token telemetry is mapped at the bridge. Codex rollout/event reconstruction now lives in `codex-core::epiphany_rollout`, where the Codex `RolloutItem` / `EventMsg` types belong. Exact retrieval path matching is native to `epiphany-core`, so `epiphany-core` has no direct vendored Codex crate dependency.
- State-update contract proof and job launch/interrupt DTO ownership now live in native `epiphany-core`. The old giant `codex_thread.rs` Epiphany state-update test fixture was cut, and `epiphany-codex-bridge` no longer imports Epiphany job launch request types from `codex_core`; vendored `CodexThread` may call the native mutation/launch law for compatibility, but it must not regain ownership of state-update semantics, heartbeat launch-plan policy, or Epiphany DTOs.
- MCP remains JSON because MCP is a JSON protocol. The latest processor cut only moved MCP refresh/OAuth/status/resource/tool-call route handling into `codex_message_processor/mcp_routes.rs`, reducing the root processor to about 8,977 lines; it did not claim MCP payloads are CultNet or make MCP an Epiphany-owned runtime contract.

Binary file not shown.

View file

@ -75,7 +75,7 @@ active_subgoals:
- The May 2026 teardown/rebuild migration in `notes/epiphany-ideal-architecture-rebuild-plan.md` is closed unless a concrete regression reopens it. Preserve the new foundation: Epiphany view/coordinator policy lives in `epiphany-core`, runtime-spine owns job/result lifecycle identity through `runtime_links`, duplicate acceptance is receipt-keyed, public findings stay typed/sealed, view expansion requires a core contract, and app-server remains routing/adaptation rather than a second policy throne.
- Treat Phase 5 update/distill/propose/promote/read/notification as complete enough unless a concrete regression appears.
- Preserve read-only boundaries for retrieve, distill, propose, freshness, context, graphQuery, roleResult, reorientResult, and the canonical view lenses for scene, jobs, roles, planning, pressure, reorient, crrc, and coordinator.
- Use `notes/epiphany-memory-graph-unified-plan.md` as the source-grounded memory architecture direction. The repo fractal dataflow graph and agent memory are one `EpiphanyMemoryGraph` substrate with repo, role, short-term, incubation, agency, candidate-intervention, and evidence profiles. Qdrant stores rebuildable semantic embeddings for typed graph documents; it is never canonical memory. The first live skeleton now exists in `epiphany-state-model` plus `epiphany-core::memory_graph`: typed graph documents, stable ids, validation, freshness propagation, context cuts, a CultCache store seam, deterministic embedding documents, manifest persistence, Qdrant rebuild through shared semantic-cache transport, and semantic context fallback that resolves cache hits back through the typed graph. Modeling graph growth now travels as typed `memoryPatchCandidates`, is applied during `roleAccept` into `state/memory-graph.msgpack`, and future role/reorientation launches prefer that persisted graph for bounded typed `memoryContext` cuts; accepted thread graphs are only bootstrap fallback. The old `thread/epiphany/propose` read route now exposes graph growth as typed `memoryPatchCandidates` instead of replacement `graphs`; generic update/promote bridge mutations reject `patch.graphs`; and view/freshness/context/graphQuery read the persisted memory graph for loaded threads. Do not add repo scanners, sleep runners, prompt integration, or ranking shortcuts before routing them through this shared substrate.
- Use `notes/epiphany-memory-graph-unified-plan.md` as the source-grounded memory architecture direction. The repo fractal dataflow graph and agent memory are one `EpiphanyMemoryGraph` substrate with repo, role, short-term, incubation, agency, candidate-intervention, and evidence profiles. Qdrant stores rebuildable semantic embeddings for typed graph documents; it is never canonical memory. The first live skeleton now exists in `epiphany-state-model` plus `epiphany-core::memory_graph`: typed graph documents, stable ids, validation, freshness propagation, context cuts, a CultCache store seam, deterministic embedding documents, manifest persistence, Qdrant rebuild through shared semantic-cache transport, and semantic context fallback that resolves cache hits back through the typed graph. Modeling graph growth now travels as typed `memoryPatchCandidates`, is applied during `roleAccept` into `state/memory-graph.msgpack`, and future role/reorientation launches prefer that persisted graph for bounded typed `memoryContext` cuts; accepted thread graphs are only bootstrap fallback. The old `thread/epiphany/propose` read route now exposes graph growth as typed `memoryPatchCandidates` instead of replacement `graphs`; generic update/promote bridge mutations reject `patch.graphs`; and view/freshness/context/graphQuery read the persisted memory graph for loaded threads through one read-route store adapter. Do not add repo scanners, sleep runners, prompt integration, or ranking shortcuts before routing them through this shared substrate.
- Keep the landed Phase 6 coordinator, limited safe-boundary CRRC automation, and pre-compaction checkpoint intervention ready for the next bridge-equipped dogfood run.
- Preserve the Ghostlight-derived heartbeat initiative seam: every lane, including coordinator/Self and Face, has a role dossier and competes for heartbeat slots through readiness/reaction timing; idle winners ruminate instead of inventing project work, and sleep windows distill the accumulated thought pressure into reviewed self-memory or doctrine.
- Treat Ghostlight as reference lineage, not a sibling runtime to preserve. The useful organ is Epiphany-native character-loop protocol: any role can project its typed dossier plus visible stimulus into a local turn packet, with Face as the first public-surface actor.
@ -297,4 +297,4 @@ current_status:
freshness propagation, summary-versus-child context cuts, and a validated
CultCache store entry; the native `epiphany-memory-graph` binary can
status/validate/query/smoke the store without Qdrant.
next_action: Continue the memory graph build unless a more urgent Codex starvation regression appears. Repo, agent, and heartbeat profile producers now map accepted architecture/dataflow graph truth, reviewed role dossier memory/goals/values, and heartbeat cognition pressure into `EpiphanyMemoryGraph` documents; compose merges profile snapshots into one validated graph; `epiphany-memory-graph compose` writes from typed graph stores; `epiphany-memory-graph refresh` can invoke live agent-memory and heartbeat-cognition producers from their typed stores; `epiphany-core` has a native typed `EpiphanyThreadState` CultCache store seam; the bridge mirrors loaded Codex-thread Epiphany state into `state/thread-state.msgpack`; CultNet advertises logical `EpiphanyThreadState` plus `EpiphanyMemoryGraph` snapshot contracts; memory graph nodes/edges/summaries project into deterministic typed embedding documents; `epiphany-memory-graph manifest` writes the embedding manifest back to the typed graph store; `epiphany-core::semantic_cache` owns shared Qdrant/Ollama transport; `epiphany-memory-graph index` fully rebuilds Qdrant from typed graph documents while keeping the manifest in the graph store; `epiphany-memory-graph semantic-context` asks Qdrant for graph IDs only, preserves hit order, then resolves context from typed graph state with a no-cache fallback; native role/reorientation worker launches now prefer persisted `state/memory-graph.msgpack` for typed `memoryContext` and no longer carry full graph cargo on native paths; typed runtime role results can persist `memoryPatchCandidates`; `epiphany-core` can review/apply append-only candidate graph growth with lifecycle receipts; `epiphany-memory-graph review-candidate` / `apply-candidate` expose that law for typed candidate files; roleResult findings expose typed `memoryPatchCandidates` for pre-acceptance operator review; bridge Modeling `roleAccept` applies runtime-spine memory patch candidates into `state/memory-graph.msgpack` through the same graph law; roleAccept responses expose typed `memoryPatchReviews`; Modeling `statePatch.graphs` is rejected by policy/prompt/schema; `thread/epiphany/propose` now exposes graph growth as typed `memoryPatchCandidates`; generic update/promote bridge mutations reject `patch.graphs`; and `thread/epiphany/view`, freshness, context, and graphQuery now prefer the persisted memory graph for loaded threads, projecting to legacy graph-shaped response lenses only at the edge. The next clean slice is extracting the repeated memory-graph-store load adapter or moving remaining legacy graph fields behind an explicit compatibility module.
next_action: Continue the memory graph build unless a more urgent Codex starvation regression appears. Repo, agent, and heartbeat profile producers now map accepted architecture/dataflow graph truth, reviewed role dossier memory/goals/values, and heartbeat cognition pressure into `EpiphanyMemoryGraph` documents; compose merges profile snapshots into one validated graph; `epiphany-memory-graph compose` writes from typed graph stores; `epiphany-memory-graph refresh` can invoke live agent-memory and heartbeat-cognition producers from their typed stores; `epiphany-core` has a native typed `EpiphanyThreadState` CultCache store seam; the bridge mirrors loaded Codex-thread Epiphany state into `state/thread-state.msgpack`; CultNet advertises logical `EpiphanyThreadState` plus `EpiphanyMemoryGraph` snapshot contracts; memory graph nodes/edges/summaries project into deterministic typed embedding documents; `epiphany-memory-graph manifest` writes the embedding manifest back to the typed graph store; `epiphany-core::semantic_cache` owns shared Qdrant/Ollama transport; `epiphany-memory-graph index` fully rebuilds Qdrant from typed graph documents while keeping the manifest in the graph store; `epiphany-memory-graph semantic-context` asks Qdrant for graph IDs only, preserves hit order, then resolves context from typed graph state with a no-cache fallback; native role/reorientation worker launches now prefer persisted `state/memory-graph.msgpack` for typed `memoryContext` and no longer carry full graph cargo on native paths; typed runtime role results can persist `memoryPatchCandidates`; `epiphany-core` can review/apply append-only candidate graph growth with lifecycle receipts; `epiphany-memory-graph review-candidate` / `apply-candidate` expose that law for typed candidate files; roleResult findings expose typed `memoryPatchCandidates` for pre-acceptance operator review; bridge Modeling `roleAccept` applies runtime-spine memory patch candidates into `state/memory-graph.msgpack` through the same graph law; roleAccept responses expose typed `memoryPatchReviews`; Modeling `statePatch.graphs` is rejected by policy/prompt/schema; `thread/epiphany/propose` now exposes graph growth as typed `memoryPatchCandidates`; generic update/promote bridge mutations reject `patch.graphs`; and `thread/epiphany/view`, freshness, context, and graphQuery now prefer the persisted memory graph for loaded threads through one read-route store adapter, projecting to legacy graph-shaped response lenses only at the edge. The next clean slice is moving remaining legacy graph fields behind an explicit compatibility module or deleting dead graph-replacement law that no live mutation path can use.

View file

@ -1,4 +1,5 @@
use codex_app_server_protocol::*;
use codex_core::CodexThread;
use codex_protocol::ThreadId;
use epiphany_codex_bridge::invalidation::epiphany_freshness_watcher_snapshot;
use epiphany_codex_bridge::launch::EPIPHANY_REORIENT_LAUNCH_BINDING_ID;
@ -22,6 +23,7 @@ use epiphany_codex_bridge::view::map_epiphany_propose_response;
use epiphany_codex_bridge::view::map_epiphany_reorient_result_response;
use epiphany_codex_bridge::view::map_epiphany_role_result_response;
use epiphany_codex_bridge::view::map_epiphany_view_response;
use epiphany_core::EpiphanyMemoryGraphSnapshot;
use epiphany_core::load_memory_graph_snapshot;
use epiphany_core::normalize_epiphany_retrieve_query;
@ -54,26 +56,14 @@ impl CodexMessageProcessor {
let loaded_thread = self.thread_manager.get_thread(thread_uuid).await.ok();
let loaded = loaded_thread.is_some();
let memory_graph_snapshot = if let Some(loaded_thread) = loaded_thread.as_ref() {
let config_snapshot = loaded_thread.config_snapshot().await;
let store_path = memory_graph_store_path(config_snapshot.cwd.as_path());
match load_memory_graph_snapshot(&store_path) {
let memory_graph_snapshot =
match load_thread_memory_graph_snapshot(loaded_thread.as_deref()).await {
Ok(snapshot) => snapshot,
Err(err) => {
self.send_internal_error(
request_id,
format!(
"failed to load Epiphany memory graph store {}: {err}",
store_path.display()
),
)
.await;
Err(message) => {
self.send_internal_error(request_id, message).await;
return;
}
}
} else {
None
};
};
let thread = match self.read_thread_view(thread_uuid, false).await {
Ok(thread) => thread,
Err(ThreadReadViewError::InvalidRequest(message)) => {
@ -248,26 +238,14 @@ impl CodexMessageProcessor {
};
let loaded_thread = self.thread_manager.get_thread(thread_uuid).await.ok();
let memory_graph_snapshot = if let Some(loaded_thread) = loaded_thread.as_ref() {
let config_snapshot = loaded_thread.config_snapshot().await;
let store_path = memory_graph_store_path(config_snapshot.cwd.as_path());
match load_memory_graph_snapshot(&store_path) {
let memory_graph_snapshot =
match load_thread_memory_graph_snapshot(loaded_thread.as_deref()).await {
Ok(snapshot) => snapshot,
Err(err) => {
self.send_internal_error(
request_id,
format!(
"failed to load Epiphany memory graph store {}: {err}",
store_path.display()
),
)
.await;
Err(message) => {
self.send_internal_error(request_id, message).await;
return;
}
}
} else {
None
};
};
let thread = match self.read_thread_view(thread_uuid, false).await {
Ok(thread) => thread,
Err(ThreadReadViewError::InvalidRequest(message)) => {
@ -328,26 +306,14 @@ impl CodexMessageProcessor {
let loaded_thread = self.thread_manager.get_thread(thread_uuid).await.ok();
let loaded = loaded_thread.is_some();
let memory_graph_snapshot = if let Some(loaded_thread) = loaded_thread.as_ref() {
let config_snapshot = loaded_thread.config_snapshot().await;
let store_path = memory_graph_store_path(config_snapshot.cwd.as_path());
match load_memory_graph_snapshot(&store_path) {
let memory_graph_snapshot =
match load_thread_memory_graph_snapshot(loaded_thread.as_deref()).await {
Ok(snapshot) => snapshot,
Err(err) => {
self.send_internal_error(
request_id,
format!(
"failed to load Epiphany memory graph store {}: {err}",
store_path.display()
),
)
.await;
Err(message) => {
self.send_internal_error(request_id, message).await;
return;
}
}
} else {
None
};
};
let thread = match self.read_thread_view(thread_uuid, false).await {
Ok(thread) => thread,
Err(ThreadReadViewError::InvalidRequest(message)) => {
@ -387,26 +353,14 @@ impl CodexMessageProcessor {
let loaded_thread = self.thread_manager.get_thread(thread_uuid).await.ok();
let loaded = loaded_thread.is_some();
let memory_graph_snapshot = if let Some(loaded_thread) = loaded_thread.as_ref() {
let config_snapshot = loaded_thread.config_snapshot().await;
let store_path = memory_graph_store_path(config_snapshot.cwd.as_path());
match load_memory_graph_snapshot(&store_path) {
let memory_graph_snapshot =
match load_thread_memory_graph_snapshot(loaded_thread.as_deref()).await {
Ok(snapshot) => snapshot,
Err(err) => {
self.send_internal_error(
request_id,
format!(
"failed to load Epiphany memory graph store {}: {err}",
store_path.display()
),
)
.await;
Err(message) => {
self.send_internal_error(request_id, message).await;
return;
}
}
} else {
None
};
};
let thread = match self.read_thread_view(thread_uuid, false).await {
Ok(thread) => thread,
Err(ThreadReadViewError::InvalidRequest(message)) => {
@ -648,3 +602,19 @@ impl CodexMessageProcessor {
self.outgoing.send_response(request_id, response).await;
}
}
async fn load_thread_memory_graph_snapshot(
loaded_thread: Option<&CodexThread>,
) -> std::result::Result<Option<EpiphanyMemoryGraphSnapshot>, String> {
let Some(loaded_thread) = loaded_thread else {
return Ok(None);
};
let config_snapshot = loaded_thread.config_snapshot().await;
let store_path = memory_graph_store_path(config_snapshot.cwd.as_path());
load_memory_graph_snapshot(&store_path).map_err(|err| {
format!(
"failed to load Epiphany memory graph store {}: {err}",
store_path.display()
)
})
}