Skip to content

ruvnet.

Python

docs(ruvnet): record 2026-09-18 improvement evidence

rUv015725be93fa
diff --git a/docs/self-improvement/2026-09-18-evidence.json b/docs/self-improvement/2026-09-18-evidence.json
index d79f7a6..2bc738e 100644
--- a/docs/self-improvement/2026-09-18-evidence.json
+++ b/docs/self-improvement/2026-09-18-evidence.json
@@ -1,9 +1,9 @@
{
"date": "2026-09-18",
"baselineCommit": "7965670321cce42b7db65138c0a8f40ddb5d584c",
- "candidateCommit": null,
+ "candidateCommit": "919e4274f51c533c92bf1386d5d19f39d33c98df",
"finalCommit": null,
- "decision": "accept-pending-publish",
+ "decision": "accept",
"candidate": "explainable-reviewed-change-search",
"version": "0.6.0",
"snapshotId": "sha256:e210ac1827e6b13d5abfba717cf1b10c7dc7c02a410224996cd646541e31ceb2",
@@ -94,6 +94,6 @@
"sources": ["https://github.com/ruvnet/ruflo/releases/tag/v3.42.4","https://arxiv.org/abs/2603.16171","https://arxiv.org/abs/2605.06132"],
"priorForecastsExpired": []
},
- "ci": null,
+ "ci": {"run":"https://github.com/ruvnet/ruvnet/actions/runs/35334295913","status":"success","node22":"success","node24":"success"},
"dashboard": "https://ruvnet-harness-evidence.ruv.chatgpt.site"
}
diff --git a/docs/self-improvement/2026-09-18.md b/docs/self-improvement/2026-09-18.md
index 35a3eab..3ca0185 100644
--- a/docs/self-improvement/2026-09-18.md
+++ b/docs/self-improvement/2026-09-18.md
@@ -2,7 +2,7 @@
## Decision
-Accepted explainable exact-token search for reviewed upstream evidence as RuVnet entry point 0.6.0, pending publication and CI. `ruvnet search` and `ruvnet_search_changes` return exact matched terms and `rawRelevance`, defined only as the matched unique query-term fraction. They do not claim semantic similarity, ranking confidence, embeddings or learning. The companion remains offline, read only and unable to refresh or learn by itself.
+Accepted and published explainable exact-token search for reviewed upstream evidence as RuVnet entry point 0.6.0. `ruvnet search` and `ruvnet_search_changes` return exact matched terms and `rawRelevance`, defined only as the matched unique query-term fraction. They do not claim semantic similarity, ranking confidence, embeddings or learning. The companion remains offline, read only and unable to refresh or learn by itself.
## Upstream review
@@ -16,6 +16,8 @@ Fifteen alternating pairs used 10,000 discovery calls and 20 cold CLI starts per
Pinned Ruflo 3.25.6 deep code and dependency scans, npm audit and a tracked-file credential-pattern check reported zero known findings. Its secret scanner covered five files and STRIDE four, so coverage is limited. Trust boundaries now include untrusted query text, exact filters, reviewed JSON, host interpretation of `rawRelevance`, and linked evidence. Residual risks include missed synonyms, stale evidence and score overinterpretation. This is not a claim of complete security.
+[Candidate commit](https://github.com/ruvnet/ruvnet/commit/919e4274f51c533c92bf1386d5d19f39d33c98df) passed [Node 22 and 24 CI](https://github.com/ruvnet/ruvnet/actions/runs/35334295913).
+
## RuVector, WASM and GNN decision
The corpus now contains 10 catalog projects plus 38 reviewed changes: 48 sanitized public records, below the frozen 100-record gate. Exact-token retrieval is implemented and measured with no persistent storage. RuVector retrieval, WASM acceleration, GNN training and persistent learning remain unimplemented and unclaimed. The future evaluation remains exact versus vector-only, fusion, GNN reranking and shuffled-graph control, requiring at least a ten-point held-out task-success gain plus latency, cold-start, memory, artifact-size and reproducibility evidence.