.. title:: packset

:description: A pack of written claims with a review clock, not a transcript with an index.

.. raw:: html

   <div class="vi-hero">
     <div class="vi-hero-brand">
       <img class="vi-hero-mark" src="_static/mark.svg" width="64" height="64" alt="" />
       <div>
         <p class="vi-hero-name">packset</p>
         <p class="vi-hero-tag">What does this seat know, standing?</p>
       </div>
     </div>
     <p class="vi-hero-tagline">A pack of written claims with a review clock, not a transcript with an index.</p>
     <div class="vi-hero-pills">
       <span>Remember / Prefer only</span>
       <span>LMDB, one writer</span>
       <span>CLI + HTTP + MCP</span>
     </div>
     <div class="vi-hero-actions">
       <a class="vi-btn vi-btn-gold" href="getting-started.html">Get started</a>
       <a class="vi-btn vi-btn-ghost" href="reference.html">Reference</a>
     </div>
   </div>

A pack is what a seat has decided to keep: one claim per atom, written on
purpose, never mined from a transcript. Search ranks the claims by a
measured panel of scorers. A review clock brings each claim back before it
is forgotten, and a claim that is not reviewed decays in the ranking. The
writer is one daemon over one Lightning Memory-Mapped Database (LMDB) file,
and every other tool is a client.

Install
=======

.. code:: console

   $ cargo binstall --locked packset          # packset, packsetd and packset-mcp
   $ packset ensure

First minute
============

.. code:: console

   $ packset remember --workspace demo "The lexical default is BM25+. It scored 0.635 hit@1 against 0.615 for BM25."
   ee17d2a07cbf1569e54e0d80496fdecf    lesson  due 2026-09-28T16:52:09.119Z
   $ packset prefer --workspace demo "CombMNZ over RRF for fusing two ballots."
   63df91bbea2b5e4e146866db0a85b4c6    preference  due 2026-09-28T16:52:09.149Z
   $ packset search --workspace demo which fusion
   packset: workspace demo
   2.0000  preference  63df91bbea2b5e4e146866db0a85b4c6    CombMNZ over RRF for fusing two ballots.
   1.0000  lesson  ee17d2a07cbf1569e54e0d80496fdecf    The lexical default is BM25+. It scored 0.635 hit@1 against 0.615 for BM25.

Two claims, written on purpose, each with a review date a day out. A search
lists the standing choice first, then the lesson. The :doc:`tutorial <getting-started>`
goes on from these two claims to review and handover.

``packset ensure`` starts the writer when it is down. With nothing set, every
client uses that writer on the default port. The dense scorer is a separate
binary, ``packset-embed``, built against Open Neural Network Exchange (ONNX)
Runtime compiled from source.
The pack answers without it, and a sitting's recall stays lexical.

Measured
========

LongMemEval\ :sub:`S`, 470 answerable questions, session-level retrieval.
BM25+ alone is 0.914 recall@5. CombMNZ of BM25+ and multilingual-e5-large
is 0.949 recall@5 (0.889 hit@1, 0.981 recall at 10). QA on the same 470
with a local 7B reader is 0.549 against a labelled-session ceiling of
0.634. Writes only on Remember: and Prefer:. The :doc:`explanation <explanation>`
is the source for those rows.

.. toctree::
   :maxdepth: 1
   :caption: Guides
   :hidden:

   getting-started
   howto
   reference
   explanation
   architecture
   seat