Skip to content

cdeust/Cortex

v4.11.0 Feature

This release adds 3 notable features for engineering teams evaluating rollout.

Published 15d MCP Data & Storage
✓ No known CVEs patched
Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

agent-memory-system anthropic causal-inference claude claude-code claude-code-plugin
+14 more
cognitive-architecture cognitive-science episodic-memory hopfield-network llm-memory long-term-memory mcp-server memory model-context-protocol neuroscience persistent-memory predictive-coding retrieval-augmented-generation vector-db

Summary

AI summary

Tail‑fill SA mode becomes default in recall, bench updates include seal merge train metrics and reranker requirement, and invariants tests re‑pin allowed‑writer lines.

Changes in this release

Feature Low

Adds explicit write_class parameter to the remember contract (M-D2, 7.4).

Adds explicit write_class parameter to the remember contract (M-D2, 7.4).

Source: llm_adapter@2026-07-15

Confidence: high

 —
Feature Low

Sets tail-fill SA mode as benchmark‑neutral default for recall.

Sets tail-fill SA mode as benchmark‑neutral default for recall.

Source: llm_adapter@2026-07-15

Confidence: high

 —
Bugfix Medium

Fixes silent-except issues on recall/write-gate/consolidation critical paths.

Fixes silent-except issues on recall/write-gate/consolidation critical paths.

Source: llm_adapter@2026-07-15

Confidence: high

 —
Bugfix Medium

Repairs dead spread_activation_memories channel and scopes it to domain.

Repairs dead spread_activation_memories channel and scopes it to domain.

Source: llm_adapter@2026-07-15

Confidence: high

 —
Bugfix Medium

Fixes candidate-contract mismatch between WRRF and SA-injected rows in recall.

Fixes candidate-contract mismatch between WRRF and SA-injected rows in recall.

Source: llm_adapter@2026-07-15

Confidence: high

 —
Bugfix Low

Aligns source‑taxonomy filter with real DB values for reheat.

Aligns source‑taxonomy filter with real DB values for reheat.

Source: llm_adapter@2026-07-15

Confidence: high

 —
Full changelog
  • chore(release): 4.11.0
  • bench: seal merge train — LongMemEval 0.9166/0.982 (tail-fill neutral), LoCoMo 0.8014/0.9157, BEAM 0.5493 (manifest 20260711T080344Z, reranker active)
  • Merge branch 'feat/spread-activation-scoped-activation'
  • feat(recall): tail-fill SA mode as benchmark-neutral default (ADR-0054 addendum)
  • test(invariants): re-pin I2 allowed-writer lines on the merge-train tree
  • Merge branch 'feat/spread-activation-scoped-activation'
  • Merge branch 'fix/silent-except-sweep'
  • Merge branch 'feat/explicit-write-class'
  • fix(recall): candidate-contract mismatch between WRRF and SA-injected rows
  • fix(observability): silent-except sweep on recall/write-gate/consolidation critical paths
  • feat(write-class): explicit write_class parameter at the remember contract (M-D2, 7.4)
  • fix(recall): repair dead spread_activation_memories channel + domain-scope it (ADR-0054)
  • docs(campaigns): INC7.2 re-heat control measurement (dry-run, read-only)
  • chore(bench): require the reranker on production-parity-claiming harnesses
  • fix(reheat): align source-taxonomy filter with real DB values (INC7.2)

Full Changelog: https://github.com/cdeust/Cortex/compare/v4.10.0...v4.11.0

View diff on GitHub

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Share on X Share on Bluesky

Track cdeust/Cortex

Get notified when new releases ship.

Sign up free

About cdeust/Cortex

Persistent memory for Claude Code grounded in computational neuroscience (41 cited papers)

All releases →

Related context

Beta — feedback welcome: [email protected]