tulving 0.0.1__py3-none-any.whl

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.
tulving/__init__.py ADDED
@@ -0,0 +1,12 @@
1
+ """Tulving — the context-budget engine for AI agents.
2
+
3
+ Persistent, typed, searchable working memory for a single AI agent, with
4
+ token-budget context curation (``curate(query, token_budget)``) as the
5
+ headline primitive. Named after Endel Tulving, the psychologist who
6
+ established that memory has types.
7
+
8
+ This is a name-holding pre-release (0.0.1). The v0.1 implementation is in
9
+ active development.
10
+ """
11
+
12
+ __version__ = "0.0.1"
@@ -0,0 +1,45 @@
1
+ Metadata-Version: 2.4
2
+ Name: tulving
3
+ Version: 0.0.1
4
+ Summary: The context-budget engine for AI agents — persistent typed memory with token-budget curation. Name-holding pre-release; v0.1 in active development.
5
+ Project-URL: Author, https://github.com/govanxa
6
+ Author-email: Abraham <abraham@fscoder.dev>
7
+ Keywords: agents,ai,context,curation,llm,mcp,memory
8
+ Classifier: Development Status :: 1 - Planning
9
+ Classifier: Intended Audience :: Developers
10
+ Classifier: Programming Language :: Python :: 3
11
+ Classifier: Programming Language :: Python :: 3.11
12
+ Classifier: Programming Language :: Python :: 3.12
13
+ Classifier: Programming Language :: Python :: 3.13
14
+ Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
15
+ Requires-Python: >=3.11
16
+ Description-Content-Type: text/markdown
17
+
18
+ # Tulving
19
+
20
+ > *The context-budget engine for AI agents.*
21
+
22
+ **Status: pre-release (0.0.1) — name-holding release; v0.1 is in active development.**
23
+
24
+ Tulving is a model-agnostic Python SDK that gives an AI agent persistent, structured, searchable working memory — and, as its headline capability, curates that memory back into a fixed token budget:
25
+
26
+ ```python
27
+ ctx = memory.curate("resuming work on the auth refactor", token_budget=4000)
28
+ ```
29
+
30
+ Named after **Endel Tulving**, the psychologist who established that memory has types (episodic vs semantic). Typed memories — facts, decisions, observations, plans, summaries, each with its own lifecycle — are Tulving's core data model.
31
+
32
+ ## What v0.1 will ship
33
+
34
+ - **Core memory engine** — store / get / semantic search over typed memories; zero infrastructure (one SQLite file, local vectors optional, no API key required)
35
+ - **Token-budget context curation** — `curate(query, token_budget)` selects, ranks, and trims memories into a prompt-ready block; `orient` mode for cold starts
36
+ - **Lazy decay & eviction** — importance fades per-type over time; decisions never decay; nothing is destroyed, only archived
37
+ - **Sessions** — per-agent session tracking, end-of-session summarization, abandoned-session recovery
38
+ - **MCP server** — six tools for Claude Code and other MCP clients, with single-writer safety
39
+ - **Adapters & export** — pluggable embeddings/LLM, JSON round-trip export
40
+
41
+ By the same author: [Kairos](https://github.com/govanxa/kairos) (`kairos-ai`) — contract-enforced AI workflows. A `kairos-ai-tulving` integration plugin is planned.
42
+
43
+ ## License
44
+
45
+ TBD (MIT or Apache 2.0) — will be finalized with the v0.1 release.
@@ -0,0 +1,4 @@
1
+ tulving/__init__.py,sha256=kNL_Ww2HXoZDA8YKjGfoZC_Ck_8Wp2X1-aGIdSaXhis,426
2
+ tulving-0.0.1.dist-info/METADATA,sha256=tWZ5WmZIv5-_M2Drf2KyD_7y-OX7qsDBgplKz_jo6mY,2470
3
+ tulving-0.0.1.dist-info/WHEEL,sha256=mffPy8wBnZQn2VnJUU5jE99KsxaSfiyMHV9Yt0aLVxs,87
4
+ tulving-0.0.1.dist-info/RECORD,,
@@ -0,0 +1,4 @@
1
+ Wheel-Version: 1.0
2
+ Generator: hatchling 1.30.1
3
+ Root-Is-Purelib: true
4
+ Tag: py3-none-any