release: v3.10.0 — Cowork compat + TDQS schema enrichment

Highlights since v3.9.0:

- feat: automatic SQLite fallback when DATABASE_URL is unset, so Cortex
  boots cleanly in Claude Cowork / Glama inspector sandboxes
- feat: CORTEX_RUNTIME env var declared in MCP env schema
- chore(tools): TDQS enrichment of all 40 MCP tool schemas — multi-sentence
  descriptions, per-parameter docs, defaults, enums, examples, bounds
- fix: SqliteMemoryStore now has memory_entities join table + mixin methods
- security: closed 2 CodeQL alerts (ReDoS in YAML frontmatter parser,
  HTTP response splitting in export filename sanitizer)
- docs: Glama score badge in README; marketplace runtime updated to
  ["cli", "cowork"]

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: cdeust

.claude-plugin/marketplace.json

@@ -6,14 +6,14 @@
   },
   "metadata": {
     "description": "Persistent memory and cognitive profiling plugins for Claude Code",
-    "version": "3.9.0"
+    "version": "3.10.0"
   },
   "plugins": [
     {
       "name": "cortex",
       "source": "./",
-      "description": "Persistent memory and cognitive profiling for Claude Code — thermodynamic memory with heat/decay, intent-aware retrieval, biological plasticity, codebase intelligence, and cognitive profiling. 33 MCP tools. Docker image available with PostgreSQL + pgvector pre-configured.",
-      "version": "3.9.0",
+      "description": "Persistent memory and cognitive profiling for Claude Code — thermodynamic memory with heat/decay, intent-aware retrieval, biological plasticity, codebase intelligence, and cognitive profiling. 40 MCP tools with enriched schemas. PostgreSQL + pgvector in CLI mode; automatic SQLite fallback in Cowork/sandboxed mode. Docker image available.",
+      "version": "3.10.0",
       "author": {
         "name": "Clement Deust",
         "email": "admin@ai-architect.tools"
@@ -23,8 +23,8 @@
       "license": "MIT",
       "keywords": ["memory", "cognitive-profiling", "mcp", "claude-code", "cortex", "knowledge-graph", "codebase-analysis", "docker"],
       "category": "productivity",
-      "runtime": ["cli"],
-      "runtime_notes": "Requires PostgreSQL 15+ with pgvector. Not compatible with Claude Cowork sandboxed environment."
+      "runtime": ["cli", "cowork"],
+      "runtime_notes": "CLI mode requires PostgreSQL 15+ with pgvector. Cowork/sandboxed mode falls back to SQLite automatically (set CORTEX_RUNTIME=cowork or leave DATABASE_URL unset)."
     }
   ]
 }

.claude-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "name": "cortex",
   "description": "Persistent memory for Claude Code — remembers across sessions automatically. Install and forget. Scientific retrieval backed by 41 published papers.",
-  "version": "3.9.1",
+  "version": "3.10.0",
   "author": {
     "name": "Clement Deust",
     "email": "admin@ai-architect.tools"

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "neuro-cortex-memory"
-version = "3.9.1"
+version = "3.10.0"
 description = "Scientifically-grounded memory system based on computational neuroscience research"
 readme = "README.md"
 license = "MIT"