superlocalmemory 3.0.1 → 3.0.2

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/CHANGELOG.md CHANGED
@@ -1,6 +1,6 @@
1
1
  # Changelog
2
2
 
3
- All notable changes to SuperLocalMemory V2 will be documented in this file.
3
+ All notable changes to SuperLocalMemory V3 will be documented in this file.
4
4
 
5
5
  The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
6
6
  and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
@@ -12,7 +12,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
12
12
  **Varun Pratap Bhardwaj**
13
13
  *Solution Architect*
14
14
 
15
- SuperLocalMemory V2 - Intelligent local memory system for AI coding assistants.
15
+ SuperLocalMemory V3 - Intelligent local memory system for AI coding assistants.
16
16
 
17
17
  ---
18
18
 
@@ -302,7 +302,7 @@ SuperLocalMemory now works across 16+ IDEs and CLI tools.
302
302
 
303
303
  ### Initial Release — Complete Rewrite
304
304
 
305
- SuperLocalMemory V2 represents a complete architectural rewrite with intelligent knowledge graphs, pattern learning, and enhanced organization.
305
+ SuperLocalMemory V3 represents a complete architectural rewrite with intelligent knowledge graphs, pattern learning, and enhanced organization.
306
306
 
307
307
  ### Added
308
308
  - **4-Layer Architecture** — Storage, Hierarchical Index, Knowledge Graph, Pattern Learning
@@ -329,7 +329,7 @@ We use [Semantic Versioning](https://semver.org/):
329
329
 
330
330
  ## License
331
331
 
332
- SuperLocalMemory V2 is released under the [MIT License](LICENSE).
332
+ SuperLocalMemory V3 is released under the [MIT License](LICENSE).
333
333
 
334
334
  ---
335
335
 
@@ -282,7 +282,7 @@ var lastFocusedElement = null; // For modal focus return
282
282
 
283
283
  If you encounter accessibility issues, please report them:
284
284
 
285
- - **GitHub Issues:** https://github.com/varun369/SuperLocalMemoryV2/issues
285
+ - **GitHub Issues:** https://github.com/qualixar/superlocalmemory/issues
286
286
  - **Label:** Use `accessibility` tag
287
287
  - **Include:** Browser, screen reader (if applicable), and steps to reproduce
288
288
 
@@ -867,8 +867,8 @@ SuperLocalMemory V2's architecture provides:
867
867
  - [CLI Commands Reference](CLI-COMMANDS-REFERENCE.md)
868
868
  - [MCP Troubleshooting](MCP-TROUBLESHOOTING.md)
869
869
  - [Universal Integration Guide](UNIVERSAL-INTEGRATION.md)
870
- - [Wiki: Knowledge Graph Guide](https://github.com/varun369/SuperLocalMemoryV2/wiki/Knowledge-Graph-Guide)
871
- - [Wiki: Pattern Learning](https://github.com/varun369/SuperLocalMemoryV2/wiki/Pattern-Learning-Explained)
870
+ - [Wiki: Knowledge Graph Guide](https://github.com/qualixar/superlocalmemory/wiki/Knowledge-Graph-Guide)
871
+ - [Wiki: Pattern Learning](https://github.com/qualixar/superlocalmemory/wiki/Pattern-Learning-Explained)
872
872
 
873
873
  ---
874
874
 
@@ -287,8 +287,8 @@ superlocalmemoryv2-profile create work
287
287
 
288
288
  ## Learn More
289
289
 
290
- - **[LangChain Wiki Guide](https://github.com/varun369/SuperLocalMemoryV2/wiki/LangChain-Integration)** — Full integration tutorial
291
- - **[LlamaIndex Wiki Guide](https://github.com/varun369/SuperLocalMemoryV2/wiki/LlamaIndex-Integration)** — Complete setup guide
290
+ - **[LangChain Wiki Guide](https://github.com/qualixar/superlocalmemory/wiki/LangChain-Integration)** — Full integration tutorial
291
+ - **[LlamaIndex Wiki Guide](https://github.com/qualixar/superlocalmemory/wiki/LlamaIndex-Integration)** — Complete setup guide
292
292
  - **[API Reference](API-REFERENCE.md)** — Python API documentation
293
293
  - **[Profiles Guide](PROFILES-GUIDE.md)** — Multi-context management
294
294
 
@@ -764,8 +764,8 @@ Plus **2 MCP prompts** and **6 MCP resources** for advanced use.
764
764
 
765
765
  1. **Check logs:** Most IDEs have MCP server logs in their developer tools
766
766
  2. **Test manually:** Run `python3 ~/.claude-memory/mcp_server.py` to see errors
767
- 3. **GitHub Issues:** https://github.com/varun369/SuperLocalMemoryV2/issues
768
- 4. **Documentation:** https://github.com/varun369/SuperLocalMemoryV2
767
+ 3. **GitHub Issues:** https://github.com/qualixar/superlocalmemory/issues
768
+ 4. **Documentation:** https://github.com/qualixar/superlocalmemory
769
769
 
770
770
  ---
771
771
 
@@ -119,7 +119,7 @@ ls -la ~/.claude-memory/pattern_learner.py
119
119
  **Fix:**
120
120
  1. Verify installation:
121
121
  ```bash
122
- cd ~/path/to/SuperLocalMemoryV2
122
+ cd ~/path/to/superlocalmemory
123
123
  ./install.sh
124
124
  ```
125
125
 
@@ -298,7 +298,7 @@ ls -la ~/.claude-memory/*.py
298
298
 
299
299
  **If files missing:**
300
300
  ```bash
301
- cd ~/path/to/SuperLocalMemoryV2-repo
301
+ cd ~/path/to/superlocalmemory-repo
302
302
  ./install.sh
303
303
  ```
304
304
 
@@ -714,7 +714,7 @@ If auto-configuration fails completely, manually add to your IDE's config:
714
714
  2. ✅ **Test server manually:** `python3 ~/.claude-memory/mcp_server.py`
715
715
  3. ✅ **Verify config is valid JSON**
716
716
  4. ✅ **Try manual configuration** (see above)
717
- 5. ✅ **Search existing issues:** https://github.com/varun369/SuperLocalMemoryV2/issues
717
+ 5. ✅ **Search existing issues:** https://github.com/qualixar/superlocalmemory/issues
718
718
 
719
719
  ### Opening an Issue
720
720
 
@@ -744,6 +744,6 @@ MIT License - See [LICENSE](../LICENSE) file
744
744
 
745
745
  ---
746
746
 
747
- **Project:** [SuperLocalMemory V2](https://github.com/varun369/SuperLocalMemoryV2)
748
- **Documentation:** [Full Docs](https://github.com/varun369/SuperLocalMemoryV2/wiki)
749
- **Issues:** [Report Issues](https://github.com/varun369/SuperLocalMemoryV2/issues)
747
+ **Project:** [SuperLocalMemory V2](https://github.com/qualixar/superlocalmemory)
748
+ **Documentation:** [Full Docs](https://github.com/qualixar/superlocalmemory/wiki)
749
+ **Issues:** [Report Issues](https://github.com/qualixar/superlocalmemory/issues)
@@ -494,7 +494,7 @@ pip install scikit-learn numpy
494
494
  - [API Reference](API-REFERENCE.md)
495
495
  - [Main README](../README.md)
496
496
 
497
- **Issues:** [GitHub Issues](https://github.com/varun369/SuperLocalMemoryV2/issues)
497
+ **Issues:** [GitHub Issues](https://github.com/qualixar/superlocalmemory/issues)
498
498
 
499
499
  ---
500
500
 
@@ -16,8 +16,8 @@ SuperLocalMemory V2 now works across **16+ IDEs and CLI tools** with automatic d
16
16
 
17
17
  ### Step 1: Install
18
18
  ```bash
19
- git clone https://github.com/varun369/SuperLocalMemoryV2.git
20
- cd SuperLocalMemoryV2
19
+ git clone https://github.com/qualixar/superlocalmemory.git
20
+ cd superlocalmemory
21
21
  ./install.sh
22
22
  ```
23
23
 
@@ -481,7 +481,7 @@ python3 ~/.claude-memory/mcp_server.py --transport http --port 8001
481
481
 
482
482
  ---
483
483
 
484
- **Questions?** Open an issue: https://github.com/varun369/SuperLocalMemoryV2/issues
484
+ **Questions?** Open an issue: https://github.com/qualixar/superlocalmemory/issues
485
485
 
486
486
  **Version:** 2.4.1
487
487
  **Author:** Varun Pratap Bhardwaj
@@ -656,8 +656,8 @@ index = HNSWIndex(
656
656
  ## Support
657
657
 
658
658
  For issues, questions, or contributions:
659
- - **GitHub Issues:** https://github.com/varun369/SuperLocalMemoryV2/issues
660
- - **Documentation:** https://github.com/varun369/SuperLocalMemoryV2/wiki
659
+ - **GitHub Issues:** https://github.com/qualixar/superlocalmemory/issues
660
+ - **Documentation:** https://github.com/qualixar/superlocalmemory/wiki
661
661
 
662
662
  ---
663
663
 
@@ -1,13 +1,13 @@
1
1
  # langchain-superlocalmemory
2
2
 
3
- LangChain chat message history backed by [SuperLocalMemory V2](https://github.com/varun369/SuperLocalMemoryV2) -- 100% local, zero cloud.
3
+ LangChain chat message history backed by [SuperLocalMemory V2](https://github.com/qualixar/superlocalmemory) -- 100% local, zero cloud.
4
4
 
5
5
  Every message stays on your machine in a SQLite database. No API keys, no subscriptions, no telemetry.
6
6
 
7
7
  ## Prerequisites
8
8
 
9
9
  - Python 3.10+
10
- - [SuperLocalMemory V2](https://github.com/varun369/SuperLocalMemoryV2) installed (`~/.claude-memory/` must exist)
10
+ - [SuperLocalMemory V2](https://github.com/qualixar/superlocalmemory) installed (`~/.claude-memory/` must exist)
11
11
  - `langchain-core >= 1.0.0`
12
12
 
13
13
  ## Installation
@@ -101,6 +101,6 @@ MIT -- see [LICENSE](../../LICENSE) for details.
101
101
 
102
102
  ## Links
103
103
 
104
- - [SuperLocalMemory V2 Repository](https://github.com/varun369/SuperLocalMemoryV2)
104
+ - [SuperLocalMemory V2 Repository](https://github.com/qualixar/superlocalmemory)
105
105
  - [Documentation](https://superlocalmemory.com/)
106
106
  - [LangChain Documentation](https://python.langchain.com/)
@@ -61,7 +61,7 @@ def _ensure_slm_imported():
61
61
  except ImportError as exc:
62
62
  raise ImportError(
63
63
  "SuperLocalMemory V2 is not installed. "
64
- "Run the installer from https://github.com/varun369/SuperLocalMemoryV2 "
64
+ "Run the installer from https://github.com/qualixar/superlocalmemory "
65
65
  "or ensure ~/.claude-memory/memory_store_v2.py exists."
66
66
  ) from exc
67
67
 
@@ -1,6 +1,6 @@
1
1
  # LlamaIndex Chat Store — SuperLocalMemory V2
2
2
 
3
- A LlamaIndex `BaseChatStore` implementation backed by [SuperLocalMemory V2](https://github.com/varun369/SuperLocalMemoryV2). All chat history stays **100% local** on your machine — zero cloud calls, zero telemetry, zero API keys.
3
+ A LlamaIndex `BaseChatStore` implementation backed by [SuperLocalMemory V2](https://github.com/qualixar/superlocalmemory). All chat history stays **100% local** on your machine — zero cloud calls, zero telemetry, zero API keys.
4
4
 
5
5
  ## Prerequisites
6
6
 
@@ -9,7 +9,7 @@ A LlamaIndex `BaseChatStore` implementation backed by [SuperLocalMemory V2](http
9
9
 
10
10
  ```bash
11
11
  # Install SuperLocalMemory V2 (one-time)
12
- curl -fsSL https://raw.githubusercontent.com/varun369/SuperLocalMemoryV2/main/install.sh | bash
12
+ curl -fsSL https://raw.githubusercontent.com/qualixar/superlocalmemory/main/install.sh | bash
13
13
  ```
14
14
 
15
15
  ## Installation
@@ -76,6 +76,6 @@ chat_store = SuperLocalMemoryChatStore(db_path="/path/to/custom/memory.db")
76
76
 
77
77
  ## Links
78
78
 
79
- - [SuperLocalMemory V2](https://github.com/varun369/SuperLocalMemoryV2)
79
+ - [SuperLocalMemory V2](https://github.com/qualixar/superlocalmemory)
80
80
  - [LlamaIndex Documentation](https://docs.llamaindex.ai/)
81
81
  - [LlamaIndex Chat Stores Guide](https://docs.llamaindex.ai/en/stable/module_guides/storing/chat_stores/)
@@ -39,8 +39,8 @@ try:
39
39
  except ImportError as exc:
40
40
  raise ImportError(
41
41
  "SuperLocalMemory V2 is not installed. "
42
- "Run: curl -fsSL https://raw.githubusercontent.com/varun369/SuperLocalMemoryV2/main/install.sh | bash\n"
43
- "Or visit: https://github.com/varun369/SuperLocalMemoryV2"
42
+ "Run: curl -fsSL https://raw.githubusercontent.com/qualixar/superlocalmemory/main/install.sh | bash\n"
43
+ "Or visit: https://github.com/qualixar/superlocalmemory"
44
44
  ) from exc
45
45
 
46
46
 
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "superlocalmemory",
3
- "version": "3.0.1",
3
+ "version": "3.0.2",
4
4
  "description": "Information-geometric agent memory with mathematical guarantees. 4-channel retrieval, Fisher-Rao similarity, zero-LLM mode, EU AI Act compliant. Works with Claude, Cursor, Windsurf, and 17+ AI tools.",
5
5
  "keywords": [
6
6
  "ai-memory",
@@ -1,126 +1,60 @@
1
1
  #!/usr/bin/env node
2
2
  /**
3
- * SuperLocalMemory V2 - NPM Postinstall Script
3
+ * SuperLocalMemory V3 - NPM Postinstall Script
4
4
  *
5
- * Copyright (c) 2026 Varun Pratap Bhardwaj
6
- * Solution Architect & Original Creator
7
- *
8
- * Licensed under MIT License (see LICENSE file)
9
- * Repository: https://github.com/varun369/SuperLocalMemoryV2
10
- *
11
- * ATTRIBUTION REQUIRED: This notice must be preserved in all copies.
5
+ * Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
6
+ * Licensed under MIT License
7
+ * Repository: https://github.com/qualixar/superlocalmemory
12
8
  */
13
9
 
14
- const { spawnSync } = require('child_process');
15
10
  const path = require('path');
16
11
  const os = require('os');
17
12
  const fs = require('fs');
18
13
 
19
14
  console.log('\n════════════════════════════════════════════════════════════');
20
- console.log(' SuperLocalMemory V2 - Post-Installation');
21
- console.log(' by Varun Pratap Bhardwaj');
22
- console.log(' https://github.com/varun369/SuperLocalMemoryV2');
15
+ console.log(' SuperLocalMemory V3 - Post-Installation');
16
+ console.log(' by Varun Pratap Bhardwaj / Qualixar');
17
+ console.log(' https://github.com/qualixar/superlocalmemory');
23
18
  console.log('════════════════════════════════════════════════════════════\n');
24
19
 
25
- // Detect if this is a global install
26
- const isGlobal = process.env.npm_config_global === 'true' ||
27
- process.env.npm_config_global === true ||
28
- (process.env.npm_config_prefix && !process.env.npm_config_prefix.includes('node_modules'));
29
-
30
- if (!isGlobal) {
31
- console.log('📦 Local installation detected.');
32
- console.log(' SuperLocalMemory is designed for global installation.');
33
- console.log(' Run: npm install -g superlocalmemory');
34
- console.log('');
35
- console.log('⏩ Skipping system installation for local install.');
36
- console.log('');
37
- process.exit(0);
38
- }
39
-
40
- console.log('🌍 Global installation detected. Running system setup...\n');
41
-
42
- // Find the package root (where install.sh/install.ps1 lives)
43
- const packageRoot = path.join(__dirname, '..');
44
- const installScript = os.platform() === 'win32'
45
- ? path.join(packageRoot, 'scripts', 'install.ps1')
46
- : path.join(packageRoot, 'scripts', 'install.sh');
47
-
48
- // Check if install script exists
49
- if (!fs.existsSync(installScript)) {
50
- console.error('❌ Error: Install script not found at ' + installScript);
51
- console.error(' Package may be corrupted. Please reinstall:');
52
- console.error(' npm uninstall -g superlocalmemory');
53
- console.error(' npm install -g superlocalmemory');
54
- process.exit(1);
55
- }
56
-
57
- // Run the appropriate install script
58
- console.log('Running installer: ' + path.basename(installScript));
59
- console.log('This will:');
60
- console.log(' • Copy files to ~/.claude-memory/');
61
- console.log(' • Configure MCP for 16+ AI tools');
62
- console.log(' • Install universal skills');
63
- console.log(' • Set up CLI commands\n');
64
-
65
- let result;
20
+ // V3 data directory
21
+ const SLM_HOME = path.join(os.homedir(), '.superlocalmemory');
22
+ const V2_HOME = path.join(os.homedir(), '.claude-memory');
66
23
 
67
- if (os.platform() === 'win32') {
68
- // Windows: Run PowerShell script
69
- console.log('Platform: Windows (PowerShell)\n');
70
- result = spawnSync('powershell', [
71
- '-ExecutionPolicy', 'Bypass',
72
- '-File', installScript,
73
- '--non-interactive'
74
- ], {
75
- stdio: 'inherit',
76
- cwd: packageRoot
77
- });
24
+ // Ensure V3 data directory exists
25
+ if (!fs.existsSync(SLM_HOME)) {
26
+ fs.mkdirSync(SLM_HOME, { recursive: true });
27
+ console.log('✓ Created data directory: ' + SLM_HOME);
78
28
  } else {
79
- // Mac/Linux: Run bash script
80
- console.log('Platform: ' + (os.platform() === 'darwin' ? 'macOS' : 'Linux') + ' (Bash)\n');
81
- result = spawnSync('bash', [installScript, '--non-interactive'], {
82
- stdio: 'inherit',
83
- cwd: packageRoot
84
- });
29
+ console.log('✓ Data directory exists: ' + SLM_HOME);
85
30
  }
86
31
 
87
- if (result.error) {
88
- console.error('\n❌ Installation failed with error:', result.error.message);
89
- console.error('\nPlease run the install script manually:');
90
- if (os.platform() === 'win32') {
91
- console.error(' powershell -ExecutionPolicy Bypass -File "' + installScript + '"');
92
- } else {
93
- console.error(' bash "' + installScript + '"');
94
- }
95
- process.exit(1);
96
- }
97
-
98
- if (result.status !== 0) {
99
- console.error('\n❌ Installation script exited with code ' + result.status);
100
- console.error('\nPlease run the install script manually:');
101
- if (os.platform() === 'win32') {
102
- console.error(' powershell -ExecutionPolicy Bypass -File "' + installScript + '"');
103
- } else {
104
- console.error(' bash "' + installScript + '"');
105
- }
106
- process.exit(result.status);
32
+ // Detect V2 installation
33
+ if (fs.existsSync(V2_HOME) && fs.existsSync(path.join(V2_HOME, 'memory.db'))) {
34
+ console.log('');
35
+ console.log('╔══════════════════════════════════════════════════════════╗');
36
+ console.log(' V2 Installation Detected ║');
37
+ console.log('╚══════════════════════════════════════════════════════════╝');
38
+ console.log('');
39
+ console.log(' Found V2 data at: ' + V2_HOME);
40
+ console.log(' Your memories are safe and will NOT be deleted.');
41
+ console.log('');
42
+ console.log(' To migrate V2 data to V3, run:');
43
+ console.log(' slm migrate');
44
+ console.log('');
45
+ console.log(' Read the migration guide:');
46
+ console.log(' https://github.com/qualixar/superlocalmemory/wiki/Migration-from-V2');
47
+ console.log('');
107
48
  }
108
49
 
109
- console.log('\n════════════════════════════════════════════════════════════');
110
- console.log(' SuperLocalMemory V2 installed successfully!');
50
+ console.log('════════════════════════════════════════════════════════════');
51
+ console.log(' SuperLocalMemory V3 installed successfully!');
52
+ console.log('');
53
+ console.log(' Quick start:');
54
+ console.log(' slm setup # First-time configuration');
55
+ console.log(' slm status # Check system status');
56
+ console.log(' slm remember "..." # Store a memory');
57
+ console.log(' slm recall "..." # Search memories');
58
+ console.log('');
59
+ console.log(' Documentation: https://github.com/qualixar/superlocalmemory/wiki');
111
60
  console.log('════════════════════════════════════════════════════════════\n');
112
-
113
- console.log('Quick Start:');
114
- console.log(' slm remember "Your first memory"');
115
- console.log(' slm recall "search query"');
116
- console.log(' slm status');
117
- console.log(' slm help\n');
118
-
119
- console.log('Documentation:');
120
- console.log(' README: https://github.com/varun369/SuperLocalMemoryV2');
121
- console.log(' Wiki: https://github.com/varun369/SuperLocalMemoryV2/wiki');
122
- console.log(' Local: ~/.claude-memory/\n');
123
-
124
- console.log('MCP Integration:');
125
- console.log(' Auto-configured for: Claude Desktop, Cursor, Windsurf, etc.');
126
- console.log(' Restart your AI tool to activate.\n');
@@ -6,7 +6,7 @@
6
6
  * Solution Architect & Original Creator
7
7
  *
8
8
  * Licensed under MIT License (see LICENSE file)
9
- * Repository: https://github.com/varun369/SuperLocalMemoryV2
9
+ * Repository: https://github.com/qualixar/superlocalmemory
10
10
  *
11
11
  * ATTRIBUTION REQUIRED: This notice must be preserved in all copies.
12
12
  */
@@ -0,0 +1,304 @@
1
+ Metadata-Version: 2.4
2
+ Name: superlocalmemory
3
+ Version: 3.0.0
4
+ Summary: Information-geometric agent memory with mathematical guarantees
5
+ Author-email: Varun Pratap Bhardwaj <admin@superlocalmemory.com>
6
+ License: MIT
7
+ Project-URL: Homepage, https://superlocalmemory.com
8
+ Project-URL: Repository, https://github.com/qualixar/superlocalmemory
9
+ Project-URL: Documentation, https://github.com/qualixar/superlocalmemory/wiki
10
+ Project-URL: Issues, https://github.com/qualixar/superlocalmemory/issues
11
+ Keywords: ai-memory,mcp-server,local-first,agent-memory,information-geometry,privacy-first,eu-ai-act
12
+ Requires-Python: >=3.11
13
+ Description-Content-Type: text/markdown
14
+ License-File: LICENSE
15
+ License-File: NOTICE
16
+ License-File: AUTHORS.md
17
+ Requires-Dist: httpx>=0.24.0
18
+ Requires-Dist: numpy<3.0.0,>=1.26.0
19
+ Requires-Dist: scipy<2.0.0,>=1.12.0
20
+ Requires-Dist: networkx>=3.0
21
+ Requires-Dist: mcp>=1.0.0
22
+ Requires-Dist: python-dateutil>=2.9.0.post0
23
+ Requires-Dist: rank-bm25>=0.2.2
24
+ Requires-Dist: vadersentiment>=3.3.2
25
+ Provides-Extra: search
26
+ Requires-Dist: sentence-transformers<4.0.0,>=2.5.0; extra == "search"
27
+ Requires-Dist: torch>=2.2.0; extra == "search"
28
+ Requires-Dist: scikit-learn<2.0.0,>=1.3.0; extra == "search"
29
+ Requires-Dist: geoopt>=0.5.0; extra == "search"
30
+ Provides-Extra: ui
31
+ Requires-Dist: fastapi<1.0.0,>=0.109.0; extra == "ui"
32
+ Requires-Dist: uvicorn[standard]<1.0.0,>=0.27.0; extra == "ui"
33
+ Requires-Dist: python-multipart<1.0.0,>=0.0.6; extra == "ui"
34
+ Provides-Extra: learning
35
+ Requires-Dist: lightgbm<5.0.0,>=4.0.0; extra == "learning"
36
+ Provides-Extra: performance
37
+ Requires-Dist: diskcache<6.0.0,>=5.6.0; extra == "performance"
38
+ Requires-Dist: orjson<4.0.0,>=3.9.0; extra == "performance"
39
+ Provides-Extra: full
40
+ Requires-Dist: superlocalmemory[learning,performance,search,ui]; extra == "full"
41
+ Provides-Extra: dev
42
+ Requires-Dist: pytest>=8.0; extra == "dev"
43
+ Requires-Dist: pytest-cov>=4.1; extra == "dev"
44
+ Dynamic: license-file
45
+
46
+ <p align="center">
47
+ <img src="https://superlocalmemory.com/assets/logo-mark.png" alt="SuperLocalMemory" width="200"/>
48
+ </p>
49
+
50
+ <h1 align="center">SuperLocalMemory V3</h1>
51
+ <p align="center"><strong>Information-Geometric Agent Memory with Mathematical Guarantees</strong></p>
52
+
53
+ <p align="center">
54
+ The first agent memory system with mathematically grounded retrieval, lifecycle management, and consistency verification. Four-channel hybrid retrieval. Three operating modes. EU AI Act compliant.
55
+ </p>
56
+
57
+ <p align="center">
58
+ <a href="https://superlocalmemory.com"><img src="https://img.shields.io/badge/Website-superlocalmemory.com-ff6b35?style=for-the-badge" alt="Website"/></a>
59
+ <a href="https://arxiv.org/abs/2603.02240"><img src="https://img.shields.io/badge/arXiv-2603.02240-b31b1b?style=for-the-badge&logo=arxiv&logoColor=white" alt="arXiv Paper"/></a>
60
+ <a href="https://zenodo.org/records/19038659"><img src="https://img.shields.io/badge/DOI-10.5281%2Fzenodo.19038659-blue?style=for-the-badge&logo=doi&logoColor=white" alt="V3 DOI"/></a>
61
+ </p>
62
+
63
+ <p align="center">
64
+ <a href="https://www.python.org/downloads/"><img src="https://img.shields.io/badge/python-3.11+-3776AB?style=flat-square&logo=python&logoColor=white" alt="Python 3.11+"/></a>
65
+ <a href="LICENSE"><img src="https://img.shields.io/badge/license-MIT-green?style=flat-square" alt="MIT License"/></a>
66
+ <a href="#three-operating-modes"><img src="https://img.shields.io/badge/EU_AI_Act-Compliant-brightgreen?style=flat-square" alt="EU AI Act"/></a>
67
+ <a href="#"><img src="https://img.shields.io/badge/tests-1400+-brightgreen?style=flat-square" alt="1400+ Tests"/></a>
68
+ <a href="#"><img src="https://img.shields.io/badge/platform-Mac_|_Linux_|_Windows-blue?style=flat-square" alt="Cross Platform"/></a>
69
+ <a href="https://github.com/qualixar/superlocalmemory/wiki"><img src="https://img.shields.io/badge/Wiki-Documentation-blue?style=flat-square" alt="Wiki"/></a>
70
+ </p>
71
+
72
+ ---
73
+
74
+ ## What is SuperLocalMemory?
75
+
76
+ SuperLocalMemory gives AI assistants persistent, structured memory that survives across sessions. Unlike simple vector stores, V3 uses **information geometry** to provide mathematically grounded retrieval, automatic contradiction detection, and self-organizing memory lifecycle management.
77
+
78
+ **Works with:** Claude, Cursor, Windsurf, VS Code Copilot, Continue, Cody, ChatGPT Desktop, Gemini CLI, JetBrains, Zed, and 17+ AI tools via MCP.
79
+
80
+ > **Upgrading from V2 (2.8.6)?** V3 is a complete architectural reinvention — new mathematical engine, new retrieval pipeline, new storage schema. Your existing data is preserved but requires migration. After installing V3, run `slm migrate` to upgrade your data. Read the [Migration Guide](docs/migration-from-v2.md) before upgrading. Backup is created automatically.
81
+
82
+ ### Key Results
83
+
84
+ | Metric | Score | Context |
85
+ |:-------|:-----:|:--------|
86
+ | LoCoMo (Mode A, zero-LLM) | **62.3%** | Highest zero-LLM score. No cloud dependency. |
87
+ | LoCoMo (Mode C, full) | **~78%** | Competitive with funded systems ($10M+) |
88
+ | Math layer improvement | **+12.7pp** | Average gain from mathematical foundations |
89
+ | Multi-hop improvement | **+12pp** | 50% vs 38% (math on vs off) |
90
+
91
+ ---
92
+
93
+ ## Three Operating Modes
94
+
95
+ | Mode | What | LLM Required? | EU AI Act | Best For |
96
+ |:----:|:-----|:-------------:|:---------:|:---------|
97
+ | **A** | Local Guardian | No | Compliant | Privacy-first, air-gapped, enterprise |
98
+ | **B** | Smart Local | Local only (Ollama) | Compliant | Enhanced quality, data stays local |
99
+ | **C** | Full Power | Cloud (optional) | Partial | Maximum accuracy, research |
100
+
101
+ **Mode A** is the only agent memory that operates with **zero cloud dependency** while achieving competitive retrieval accuracy. All data stays on your device. No API keys required.
102
+
103
+ ---
104
+
105
+ ## Architecture
106
+
107
+ ```
108
+ Query ──► Strategy Classifier ──► 4 Parallel Channels:
109
+ ├── Semantic (Fisher-Rao graduated similarity)
110
+ ├── BM25 (keyword matching, k1=1.2, b=0.75)
111
+ ├── Entity Graph (spreading activation, 3 hops)
112
+ └── Temporal (date-aware retrieval)
113
+
114
+ RRF Fusion (k=60)
115
+
116
+ Scene Expansion + Bridge Discovery
117
+
118
+ Cross-Encoder Reranking
119
+
120
+ ◄── Top-K Results with channel scores
121
+ ```
122
+
123
+ ### Mathematical Foundations (Novel Contributions)
124
+
125
+ 1. **Fisher-Rao Retrieval Metric** — Similarity scoring derived from the Fisher information structure of diagonal Gaussian families. Graduated ramp from cosine to Fisher-information-weighted scoring over the first 10 accesses per memory.
126
+
127
+ 2. **Sheaf Cohomology for Consistency** — Algebraic topology detects contradictions between facts by computing coboundary norms on the knowledge graph. Non-trivial restriction maps amplify disagreements along discriminative subspaces.
128
+
129
+ 3. **Riemannian Langevin Lifecycle** — Memory positions evolve on the Poincare ball via a discretized Langevin SDE. Frequently accessed memories stay near the origin (ACTIVE); neglected memories diffuse toward the boundary (ARCHIVED). The potential is modulated by access frequency, age, and importance.
130
+
131
+ ---
132
+
133
+ ## Quick Start
134
+
135
+ ### Install via npm (recommended)
136
+
137
+ ```bash
138
+ npm install -g superlocalmemory
139
+ slm setup
140
+ ```
141
+
142
+ ### Install via pip
143
+
144
+ ```bash
145
+ pip install superlocalmemory
146
+ # or with all features:
147
+ pip install "superlocalmemory[full]"
148
+ ```
149
+
150
+ ### First Use
151
+
152
+ ```bash
153
+ # Store a memory
154
+ slm remember "Alice works at Google as a Staff Engineer"
155
+
156
+ # Recall
157
+ slm recall "What does Alice do?"
158
+
159
+ # Check status
160
+ slm status
161
+
162
+ # Switch modes
163
+ slm mode a # Zero-LLM (default)
164
+ slm mode b # Local Ollama
165
+ slm mode c # Full power
166
+ ```
167
+
168
+ ### MCP Integration (Claude, Cursor, etc.)
169
+
170
+ Add to your IDE's MCP config:
171
+
172
+ ```json
173
+ {
174
+ "mcpServers": {
175
+ "superlocalmemory": {
176
+ "command": "slm",
177
+ "args": ["mcp"]
178
+ }
179
+ }
180
+ }
181
+ ```
182
+
183
+ 24 MCP tools available: `remember`, `recall`, `search`, `fetch`, `list_recent`, `get_status`, `build_graph`, `switch_profile`, `health`, `consistency_check`, `recall_trace`, and more.
184
+
185
+ ---
186
+
187
+ ## V3 Engine Features
188
+
189
+ ### Retrieval (4-Channel Hybrid)
190
+ - Semantic similarity with Fisher-Rao information geometry
191
+ - BM25 keyword matching (persisted tokens, survives restart)
192
+ - Entity graph with spreading activation (3-hop, decay=0.7)
193
+ - Temporal date-aware retrieval with interval support
194
+ - RRF fusion (k=60) + cross-encoder reranking
195
+
196
+ ### Intelligence
197
+ - 11-step ingestion pipeline (entity resolution, fact extraction, emotional tagging, scene building, sheaf consistency)
198
+ - Adaptive learning with LightGBM-based ranking (3-phase bootstrap)
199
+ - Behavioral pattern detection (query habits, entity preferences, active hours)
200
+ - Outcome tracking for retrieval feedback loops
201
+
202
+ ### Trust & Compliance
203
+ - Bayesian Beta distribution trust scoring (per-agent, per-fact)
204
+ - Trust gates (block low-trust agents from writing/deleting)
205
+ - ABAC (Attribute-Based Access Control) with DB-persisted policies
206
+ - GDPR Article 15/17 compliance (full export + complete erasure)
207
+ - EU AI Act data sovereignty (Mode A: zero cloud, data stays local)
208
+ - Tamper-proof hash-chain audit trail (SHA-256 linked entries)
209
+ - Data provenance tracking (who created what, when, from where)
210
+
211
+ ### Infrastructure
212
+ - 17-tab web dashboard (trust visualization, math health, recall lab)
213
+ - 17+ IDE integrations with pre-built configs
214
+ - Profile isolation (16+ independent memory spaces)
215
+ - V2 to V3 migration tool (zero data loss, rollback support)
216
+ - Auto-capture and auto-recall hooks for Claude Code
217
+
218
+ ---
219
+
220
+ ## Benchmarks
221
+
222
+ Evaluated on the [LoCoMo benchmark](https://arxiv.org/abs/2402.09714) (Long Conversation Memory):
223
+
224
+ ### Mode A Ablation (conv-30, 81 questions, zero-LLM)
225
+
226
+ | Configuration | Micro Avg | Multi-Hop | Open Domain |
227
+ |:-------------|:---------:|:---------:|:-----------:|
228
+ | Full (all layers) | **62.3%** | **50%** | **78%** |
229
+ | Math layers off | 59.3% | 38% | 70% |
230
+ | Entity channel off | 56.8% | 38% | 73% |
231
+ | BM25 channel off | 53.2% | 23% | 71% |
232
+ | Cross-encoder off | 31.8% | 17% | — |
233
+
234
+ ### Competitive Landscape
235
+
236
+ | System | Score | LLM Required | Open Source | EU AI Act |
237
+ |:-------|:-----:|:------------:|:-----------:|:---------:|
238
+ | EverMemOS | 92.3% | Yes | No | No |
239
+ | MemMachine | 91.7% | Yes | No | No |
240
+ | Hindsight | 89.6% | Yes | No | No |
241
+ | **SLM V3 Mode C** | **~78%** | Optional | **Yes** | Partial |
242
+ | **SLM V3 Mode A** | **62.3%** | **No** | **Yes** | **Yes** |
243
+ | Mem0 ($24M) | 34.2% F1 | Yes | Partial | No |
244
+
245
+ *SLM V3 is the only system offering a fully local mode with mathematical guarantees and EU AI Act compliance.*
246
+
247
+ ---
248
+
249
+ ## Research Papers
250
+
251
+ ### V3: Information-Geometric Foundations
252
+ > **SuperLocalMemory V3: Information-Geometric Foundations for Zero-LLM Enterprise Agent Memory**
253
+ > Varun Pratap Bhardwaj (2026)
254
+ > [Zenodo DOI: 10.5281/zenodo.19038659](https://zenodo.org/records/19038659)
255
+
256
+ ### V2: Architecture & Engineering
257
+ > **SuperLocalMemory: A Structured Local Memory Architecture for Persistent AI Agent Context**
258
+ > Varun Pratap Bhardwaj (2026)
259
+ > [arXiv:2603.02240](https://arxiv.org/abs/2603.02240) | [Zenodo DOI: 10.5281/zenodo.18709670](https://zenodo.org/records/18709670)
260
+
261
+ ---
262
+
263
+ ## Project Structure
264
+
265
+ ```
266
+ superlocalmemory/
267
+ ├── src/superlocalmemory/ # Python package (17 sub-packages)
268
+ │ ├── core/ # Engine, config, modes, profiles
269
+ │ ├── retrieval/ # 4-channel retrieval + fusion + reranking
270
+ │ ├── math/ # Fisher-Rao, Sheaf, Langevin
271
+ │ ├── encoding/ # 11-step ingestion pipeline
272
+ │ ├── storage/ # SQLite with WAL, FTS5, migrations
273
+ │ ├── trust/ # Bayesian scoring, gates, provenance
274
+ │ ├── compliance/ # GDPR, EU AI Act, ABAC, audit chain
275
+ │ ├── learning/ # Adaptive ranking, behavioral patterns
276
+ │ ├── mcp/ # MCP server (24 tools, 6 resources)
277
+ │ ├── cli/ # CLI with setup wizard
278
+ │ └── server/ # Dashboard API + UI server
279
+ ├── tests/ # 1400+ tests
280
+ ├── ui/ # 17-tab web dashboard
281
+ ├── ide/ # IDE configs for 17+ tools
282
+ ├── docs/ # Documentation
283
+ └── pyproject.toml # Modern Python packaging
284
+ ```
285
+
286
+ ---
287
+
288
+ ## Contributing
289
+
290
+ See [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.
291
+
292
+ ## License
293
+
294
+ MIT License. See [LICENSE](LICENSE).
295
+
296
+ ## Attribution
297
+
298
+ Part of [Qualixar](https://qualixar.com) | Author: [Varun Pratap Bhardwaj](https://varunpratap.com)
299
+
300
+ ---
301
+
302
+ <p align="center">
303
+ <sub>Built with mathematical rigor. Not in the race — here to help everyone build better AI memory systems.</sub>
304
+ </p>
@@ -0,0 +1,158 @@
1
+ AUTHORS.md
2
+ LICENSE
3
+ NOTICE
4
+ README.md
5
+ pyproject.toml
6
+ src/superlocalmemory/__init__.py
7
+ src/superlocalmemory.egg-info/PKG-INFO
8
+ src/superlocalmemory.egg-info/SOURCES.txt
9
+ src/superlocalmemory.egg-info/dependency_links.txt
10
+ src/superlocalmemory.egg-info/entry_points.txt
11
+ src/superlocalmemory.egg-info/requires.txt
12
+ src/superlocalmemory.egg-info/top_level.txt
13
+ src/superlocalmemory/attribution/__init__.py
14
+ src/superlocalmemory/attribution/mathematical_dna.py
15
+ src/superlocalmemory/attribution/signer.py
16
+ src/superlocalmemory/attribution/watermark.py
17
+ src/superlocalmemory/cli/__init__.py
18
+ src/superlocalmemory/cli/commands.py
19
+ src/superlocalmemory/cli/main.py
20
+ src/superlocalmemory/cli/migrate_cmd.py
21
+ src/superlocalmemory/cli/post_install.py
22
+ src/superlocalmemory/cli/setup_wizard.py
23
+ src/superlocalmemory/compliance/__init__.py
24
+ src/superlocalmemory/compliance/abac.py
25
+ src/superlocalmemory/compliance/audit.py
26
+ src/superlocalmemory/compliance/eu_ai_act.py
27
+ src/superlocalmemory/compliance/gdpr.py
28
+ src/superlocalmemory/compliance/lifecycle.py
29
+ src/superlocalmemory/compliance/retention.py
30
+ src/superlocalmemory/compliance/scheduler.py
31
+ src/superlocalmemory/core/__init__.py
32
+ src/superlocalmemory/core/config.py
33
+ src/superlocalmemory/core/embeddings.py
34
+ src/superlocalmemory/core/engine.py
35
+ src/superlocalmemory/core/hooks.py
36
+ src/superlocalmemory/core/maintenance.py
37
+ src/superlocalmemory/core/modes.py
38
+ src/superlocalmemory/core/profiles.py
39
+ src/superlocalmemory/core/registry.py
40
+ src/superlocalmemory/dynamics/__init__.py
41
+ src/superlocalmemory/dynamics/fisher_langevin_coupling.py
42
+ src/superlocalmemory/encoding/__init__.py
43
+ src/superlocalmemory/encoding/consolidator.py
44
+ src/superlocalmemory/encoding/emotional.py
45
+ src/superlocalmemory/encoding/entity_resolver.py
46
+ src/superlocalmemory/encoding/entropy_gate.py
47
+ src/superlocalmemory/encoding/fact_extractor.py
48
+ src/superlocalmemory/encoding/foresight.py
49
+ src/superlocalmemory/encoding/graph_builder.py
50
+ src/superlocalmemory/encoding/observation_builder.py
51
+ src/superlocalmemory/encoding/scene_builder.py
52
+ src/superlocalmemory/encoding/signal_inference.py
53
+ src/superlocalmemory/encoding/temporal_parser.py
54
+ src/superlocalmemory/encoding/type_router.py
55
+ src/superlocalmemory/hooks/__init__.py
56
+ src/superlocalmemory/hooks/auto_capture.py
57
+ src/superlocalmemory/hooks/auto_recall.py
58
+ src/superlocalmemory/hooks/ide_connector.py
59
+ src/superlocalmemory/hooks/rules_engine.py
60
+ src/superlocalmemory/infra/__init__.py
61
+ src/superlocalmemory/infra/auth_middleware.py
62
+ src/superlocalmemory/infra/backup.py
63
+ src/superlocalmemory/infra/cache_manager.py
64
+ src/superlocalmemory/infra/event_bus.py
65
+ src/superlocalmemory/infra/rate_limiter.py
66
+ src/superlocalmemory/infra/webhook_dispatcher.py
67
+ src/superlocalmemory/learning/__init__.py
68
+ src/superlocalmemory/learning/adaptive.py
69
+ src/superlocalmemory/learning/behavioral.py
70
+ src/superlocalmemory/learning/behavioral_listener.py
71
+ src/superlocalmemory/learning/bootstrap.py
72
+ src/superlocalmemory/learning/cross_project.py
73
+ src/superlocalmemory/learning/database.py
74
+ src/superlocalmemory/learning/engagement.py
75
+ src/superlocalmemory/learning/features.py
76
+ src/superlocalmemory/learning/feedback.py
77
+ src/superlocalmemory/learning/outcomes.py
78
+ src/superlocalmemory/learning/project_context.py
79
+ src/superlocalmemory/learning/ranker.py
80
+ src/superlocalmemory/learning/source_quality.py
81
+ src/superlocalmemory/learning/workflows.py
82
+ src/superlocalmemory/llm/__init__.py
83
+ src/superlocalmemory/llm/backbone.py
84
+ src/superlocalmemory/math/__init__.py
85
+ src/superlocalmemory/math/fisher.py
86
+ src/superlocalmemory/math/langevin.py
87
+ src/superlocalmemory/math/sheaf.py
88
+ src/superlocalmemory/mcp/__init__.py
89
+ src/superlocalmemory/mcp/resources.py
90
+ src/superlocalmemory/mcp/server.py
91
+ src/superlocalmemory/mcp/tools.py
92
+ src/superlocalmemory/mcp/tools_core.py
93
+ src/superlocalmemory/mcp/tools_v28.py
94
+ src/superlocalmemory/mcp/tools_v3.py
95
+ src/superlocalmemory/retrieval/__init__.py
96
+ src/superlocalmemory/retrieval/agentic.py
97
+ src/superlocalmemory/retrieval/ann_index.py
98
+ src/superlocalmemory/retrieval/bm25_channel.py
99
+ src/superlocalmemory/retrieval/bridge_discovery.py
100
+ src/superlocalmemory/retrieval/engine.py
101
+ src/superlocalmemory/retrieval/entity_channel.py
102
+ src/superlocalmemory/retrieval/fusion.py
103
+ src/superlocalmemory/retrieval/profile_channel.py
104
+ src/superlocalmemory/retrieval/reranker.py
105
+ src/superlocalmemory/retrieval/semantic_channel.py
106
+ src/superlocalmemory/retrieval/strategy.py
107
+ src/superlocalmemory/retrieval/temporal_channel.py
108
+ src/superlocalmemory/server/__init__.py
109
+ src/superlocalmemory/server/api.py
110
+ src/superlocalmemory/server/security_middleware.py
111
+ src/superlocalmemory/server/ui.py
112
+ src/superlocalmemory/server/routes/__init__.py
113
+ src/superlocalmemory/server/routes/agents.py
114
+ src/superlocalmemory/server/routes/backup.py
115
+ src/superlocalmemory/server/routes/behavioral.py
116
+ src/superlocalmemory/server/routes/compliance.py
117
+ src/superlocalmemory/server/routes/data_io.py
118
+ src/superlocalmemory/server/routes/events.py
119
+ src/superlocalmemory/server/routes/helpers.py
120
+ src/superlocalmemory/server/routes/learning.py
121
+ src/superlocalmemory/server/routes/lifecycle.py
122
+ src/superlocalmemory/server/routes/memories.py
123
+ src/superlocalmemory/server/routes/profiles.py
124
+ src/superlocalmemory/server/routes/stats.py
125
+ src/superlocalmemory/server/routes/v3_api.py
126
+ src/superlocalmemory/server/routes/ws.py
127
+ src/superlocalmemory/storage/__init__.py
128
+ src/superlocalmemory/storage/access_control.py
129
+ src/superlocalmemory/storage/database.py
130
+ src/superlocalmemory/storage/migrations.py
131
+ src/superlocalmemory/storage/models.py
132
+ src/superlocalmemory/storage/schema.py
133
+ src/superlocalmemory/storage/v2_migrator.py
134
+ src/superlocalmemory/trust/__init__.py
135
+ src/superlocalmemory/trust/gate.py
136
+ src/superlocalmemory/trust/provenance.py
137
+ src/superlocalmemory/trust/scorer.py
138
+ src/superlocalmemory/trust/signals.py
139
+ tests/test_auto_hooks.py
140
+ tests/test_behavioral_full.py
141
+ tests/test_cli.py
142
+ tests/test_compliance_full.py
143
+ tests/test_config_system.py
144
+ tests/test_engine_hooks.py
145
+ tests/test_event_bus.py
146
+ tests/test_features.py
147
+ tests/test_final_locomo_mini.py
148
+ tests/test_ide_connector.py
149
+ tests/test_infra.py
150
+ tests/test_learning_advanced.py
151
+ tests/test_learning_collectors.py
152
+ tests/test_llm_provider.py
153
+ tests/test_mcp_server.py
154
+ tests/test_migration.py
155
+ tests/test_post_install.py
156
+ tests/test_ranker.py
157
+ tests/test_trust_full.py
158
+ tests/test_v3_api.py
@@ -0,0 +1,2 @@
1
+ [console_scripts]
2
+ slm = superlocalmemory.cli.main:main
@@ -0,0 +1,33 @@
1
+ httpx>=0.24.0
2
+ numpy<3.0.0,>=1.26.0
3
+ scipy<2.0.0,>=1.12.0
4
+ networkx>=3.0
5
+ mcp>=1.0.0
6
+ python-dateutil>=2.9.0.post0
7
+ rank-bm25>=0.2.2
8
+ vadersentiment>=3.3.2
9
+
10
+ [dev]
11
+ pytest>=8.0
12
+ pytest-cov>=4.1
13
+
14
+ [full]
15
+ superlocalmemory[learning,performance,search,ui]
16
+
17
+ [learning]
18
+ lightgbm<5.0.0,>=4.0.0
19
+
20
+ [performance]
21
+ diskcache<6.0.0,>=5.6.0
22
+ orjson<4.0.0,>=3.9.0
23
+
24
+ [search]
25
+ sentence-transformers<4.0.0,>=2.5.0
26
+ torch>=2.2.0
27
+ scikit-learn<2.0.0,>=1.3.0
28
+ geoopt>=0.5.0
29
+
30
+ [ui]
31
+ fastapi<1.0.0,>=0.109.0
32
+ uvicorn[standard]<1.0.0,>=0.27.0
33
+ python-multipart<1.0.0,>=0.0.6
@@ -0,0 +1 @@
1
+ superlocalmemory