andy/noema
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: add long-term memory systems with fused retrieval and memory explorer

Browse Source 
Phase 5 (ADR-0003). pkg/concepts fixes the epistemic vocabulary
(observation, fact, memory, claim, hypothesis, belief, prediction,
assumption, inference, opinion), memory kinds and relation names.

internal/memory stores episodic, semantic, procedural and
autobiographical memories in one mind-scoped table with trust-derived
credibility, importance, salience and decaying activation. Retrieval
fuses full-text rank (any-term by default), optional pgvector similarity,
recency, activation, importance and credibility, and returns the
component scores so every recall can be explained. Decay lowers
activation and never deletes; retraction and supersession keep rows for
provenance. Relationships are constrained to one mind at insert time.
Embeddings live in a conditional pgvector table keyed by model so vector
spaces are never mixed; without an embedder the system degrades to text
and graph retrieval.

Attention gates encoding: the attend handler turns admitted observations
into episodic memories (operator changes become autobiographical ones)
with memory_created events; the retrieve process proposes recalled
memories as candidates and admitted recalls are reinforced with
memory_retrieved events carrying their scoring components.

Memory explorer UI (search with component breakdown, timeline, filters,
operator knowledge entry), memory detail with provenance, relationships,
citing events and an inaccurate-memory correction; API; noemactl memory
commands; decay and embed-missing background jobs.
This commit is contained in:
Andy Dixon
2026-06-20 23:47:50 +01:00
parent e9c1b25c80
commit 03631e5685
30 changed files with 1957 additions and 25 deletions
Download Patch File Download Diff File Expand all files Collapse all files
web/static/app.css
+8
View File
@@ -160,3 +160,11 @@ ol.corr li.current { font-weight: 600; }
ul.wm li { padding: .45rem 0; border-bottom: 1px solid var(--border); } ul.wm li:last-child { border-bottom: 0; }
.weights { display: grid; grid-template-columns: repeat(auto-fill, minmax(150px, 1fr)); gap: .5rem; }
.weights label { font-size: .8rem; }
/* Memory */
.memtext { font-size: 1.1rem; max-width: 80ch; }
.comp { display: inline-block; background: var(--surface-2); border-radius: 4px; padding: 0 .35rem; margin: .1rem 0; font-family: var(--mono); font-size: .72rem; }
tr.dayrow td { background: var(--surface-2); font-weight: 600; font-size: .78rem; text-transform: uppercase; letter-spacing: .05em; color: var(--muted); }
.badge.k-episodic { background: #e0f2fe; color: #0369a1; } .badge.k-semantic { background: #ede9fe; color: #6d28d9; } .badge.k-procedural { background: #dcfce7; color: #15803d; } .badge.k-autobiographical { background: #fef3c7; color: #b45309; }
:root[data-theme="dark"] .badge.k-episodic, :root:not([data-theme="light"]) .badge.k-episodic { background: #0c2a3a; color: #7dd3fc; }
:root[data-theme="dark"] .badge.k-semantic { background: #2a1f4d; color: #c4b5fd; } :root[data-theme="dark"] .badge.k-procedural { background: #12331d; color: #86efac; } :root[data-theme="dark"] .badge.k-autobiographical { background: #3a2a0a; color: #fcd34d; }