superlocalmemory 4.0.7 → 4.0.9

package/CHANGELOG.md

@@ -5,6 +5,224 @@ All notable changes to SuperLocalMemory will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [4.0.9] — Findable the moment you write it
+
+### Fixed
+- **A memory you just wrote could not be found by asking for it.** Until a
+  background pass caught up, a new memory was reachable only by quoting its own
+  wording — which is not how anyone asks a question. "What am I working on"
+  shares no distinctive word with any particular note, so the memory describing
+  what you are doing right now was the hardest thing in the store to find. Every
+  way of storing one — command line, tool interface, dashboard — now makes it
+  searchable by meaning before it answers, bounded so that a slow or missing
+  model costs a short wait and never the write itself. The receipt says which of
+  three states a memory is in: stored, findable by its wording, findable by
+  meaning.
+- **"What am I working on" answered with notes from a month ago.** Four separate
+  faults sat between the question and the answer. The present-tense path weighted
+  word-matching more heavily than the path it replaced, so a stale note
+  containing the word "working" beat a fresh one. A recency lookup was switched
+  off by an unrelated condition that is almost always true. A fact carrying
+  several dated events crowded out other recent facts with copies of itself. And
+  the lookup drew its candidates from a list that is only assembled later, and
+  only for notes that mention a recognisable name and a date — so a plain note
+  written seconds ago was not merely ranked low, it was not a candidate at all.
+  Recent notes are now read from the memories themselves. Questions about topics
+  still answer on topic — asking about the schema does not return whatever is
+  newest.
+- **A memory could be reported as findable by meaning when it was not.** Two
+  things have to agree before a search on meaning can reach a memory, and one of
+  the two ways of storing one wrote them in the wrong order: if the second write
+  was refused — because the store was at a different size, or the search
+  extension was unavailable on the platform — the memory was left in a state that
+  reads as complete, so the pass that repairs missed memories skipped it forever
+  while the receipt claimed success. Both writes now happen in the one order that
+  is safe, through shared code rather than two copies of it, and if the second
+  fails the first is undone so the repair pass can retry. "Already findable" is
+  now decided by asking whether a search could actually return the memory.
+- **Questions like "what is the latest authentication design" returned the newest
+  notes rather than an answer.** Two words routed any question containing them
+  down the time-based path, where a question with no recognisable name in it fell
+  back to listing recent memories with no regard for the topic. Asking what is
+  latest about something now answers about that something. "What am I currently
+  focused on" still answers with current work.
+- **A question about a span of time missed anything recorded as spanning it.** An
+  entry describing a period rather than a moment carries no single date, and the
+  lookup for date-anchored questions required one — so "what happened in March
+  2024" could not return an entry that says it covers March 2024.
+- **The documented switch for turning the present-tense ranking off did nothing
+  when set in configuration.** It worked only as an environment variable, which
+  is not what it is documented as. Both now work, and the environment variable
+  still wins.
+- **Converting a store could stop making progress instead of finishing.** A row
+  that cannot be converted — malformed, or the wrong size — was left in place and
+  then selected again on the next pass, so a single bad row meant the conversion
+  ran indefinitely without converting anything further. It now moves past what it
+  cannot convert, reports how many, and finishes.
+- **Retention could delete half of a safety copy.** The two files taken before an
+  upgrade were sometimes counted as two separate copies rather than one, so
+  retention could keep one file of a pair and discard the other, leaving a copy
+  that cannot be restored from. Choosing the newest copy also relied on file
+  timestamps, which on some filesystems cannot tell two copies apart; it now uses
+  the timestamp in the name, which can.
+- **An upgrade could be refused on a disk with room for it.** The space check
+  asked for roughly twice what the copy actually needs.
+- **Retrying an upgrade that had failed did not take a fresh copy first.** A
+  failed step was treated as a finished one, so the retry ran against the
+  half-changed store with no new copy taken. The message now also says where the
+  copy from the first attempt is.
+- **Shutting the service down could exceed its shutdown budget** and be killed
+  mid-write, because the workers that make new memories searchable were never
+  released. They are now released on shutdown, and a memory arriving during
+  shutdown is declined rather than starting new work.
+- **A question about a particular time missed anything older than the most recent
+  few thousand events**, returning nothing rather than answering slowly.
+- **`slm remember --sync` promised to wait and did not.** The flag is now
+  honoured, and asking to wait buys a longer attempt at making the memory
+  searchable.
+- **A failed upgrade could report that your data was untouched when earlier steps
+  had already been applied.** The message now distinguishes a partial change from
+  no change, and points at the copy taken beforehand.
+- **The copy taken before an upgrade could be deleted by a later start.** It was
+  being retaken on every ordinary start and pruned by retention, so two restarts
+  after an upgrade removed the last copy of the original. It is now taken only
+  when something is actually about to change.
+
+### Changed
+- **Embeddings are stored in binary rather than text** — the same values to the
+  last decimal, five and a half times smaller. Existing stores are converted in
+  place; the conversion copies the store aside and verifies the copy first,
+  refuses while a service still holds it, and asks before rewriting anything.
+- **Recall tail latency is roughly halved**, and the time a write may spend
+  making itself findable has been raised rather than lowered. Giving up early on
+  that was never free: it decided whether a memory could be found by asking a
+  question or only by quoting its own words. A retrieval channel is likewise
+  given more time to answer, because a channel that runs out of time contributes
+  nothing at all and the answer silently loses whatever only it could see. That
+  limit is now what it was always described as — a guard against one part of
+  retrieval getting stuck, set far above how long any of them actually takes,
+  rather than a cutoff that trims the wait by dropping part of the answer.
+- **When part of an answer is missing, the reply now says which part.** If one
+  source of results has to be abandoned, its memories are absent from that
+  answer rather than merely late — so asking again can legitimately give
+  something better. Replies carry the list of anything that was abandoned, which
+  is empty on a healthy answer, so it is possible to tell an incomplete answer
+  from a changed one instead of guessing.
+- **Asking the same question twice gives the same answer.** This had two
+  separate causes. Several points in retrieval ordered results by score alone,
+  so memories that scored equally were ranked by where they happened to sit on
+  disk or which thread finished first. And two stages that improve an answer —
+  one that brings in related memories, one that re-scores results by how they
+  connect to the things your question mentions — ran only if the answer was not
+  already taking too long. Both stages change which memory comes first, so
+  whether they ran decided the answer, and the time they were measured against
+  was close to how long a typical question takes anyway. The same question
+  asked twice took different paths depending on how busy the machine was, and
+  nothing in the reply said so. Ordering is now decided only by the data, and
+  those stages always run. Measured over 140 questions on a 1 GB store, asked
+  twice: the top answer changed on 15% of them before and 5.7% after, and the
+  top ten changed on 41% before and 13.6% after. Answering takes roughly a
+  tenth of a second longer.
+- **A ranking adjustment for questions about time now actually affects the
+  order results come back in.** It was being calculated and then discarded.
+- **Storing a memory stays inside its time budget.** The durable write and the
+  work that makes the memory findable by meaning run one after the other, and
+  each had been given the whole budget, so a slow write could push the reply to
+  several seconds. The second now takes what the first left.
+- **Converting a store can no longer report success while storing nonsense.**
+  Two malformed shapes — a vector of one-element lists, and numbers too large to
+  represent — passed the size check and would have been written as garbage.
+  Both are now rejected and reported, and the memory keeps its original value.
+- **A memory is no longer reported findable by meaning unless both of its stored
+  forms agree.** Either one alone could be present without the other, and each
+  case was invisible to the pass that repairs it.
+- **Shutting the service down twice in one process no longer halves the capacity
+  that makes new memories findable.**
+- **More of your memories can be reached at all.** One stage of retrieval decides
+  the final set of results, and the number of candidates handed to it had been cut
+  to save time — which quietly meant anything ranked below that cut could never be
+  returned, no matter how well it matched. The cut is undone. Recall spends a
+  little longer and can reach everything.
+- **Questions about the past distinguish two days ago from a month ago again.**
+  The adjustment applied to time-based questions was capped so low that it landed
+  on the same value for everything inside about five weeks, so it could not tell
+  recent from distant. Note that this changes ranking order for such questions.
+- Storing a memory reports whether it is searchable by meaning yet, so a caller
+  no longer has to assume.
+
+## [4.0.8] — Where you were actually looking
+
+### Added
+- **Summaries reach agents and the dashboard.** 4.0.7 gave the summary layer a
+  command; an agent still had no way to ask for one, and neither did anyone
+  reading the dashboard. Adds the `get_memory_summary` tool and a **Summaries**
+  tab under Memories with Today, Yesterday, and a picker listing the projects
+  SuperLocalMemory has actually recorded activity in. Every result states how
+  many memories it was built from and how much it covered.
+- **Bounded Loops has its own page**, alongside the other integrations. It shows
+  whether Bounded Loops is installed, which agreement the two products
+  negotiated, and which finished runs have been imported — and states plainly
+  that imported evidence is read-only: SuperLocalMemory never learns from it, no
+  file paths leave your workspace, and the tamper-evidence it carries is not the
+  same thing as independent verification.
+- **Mesh shows its history**, not only who is connected this second. Sessions
+  leave the mesh when they end, so the page was empty between sessions even on a
+  machine with plenty of mesh activity.
+
+### Changed
+- **Recall Lab is the first thing you see under Memories**, with example
+  questions to start from. Summaries sits next to it. The three browsing views
+  follow.
+
+### Fixed
+- **Memory stopped learning from use.** Recall marks which memories it returned
+  so that later use can be credited back, but the mark never reached the
+  assistant, so nothing was ever credited. Every recall was scored as neither
+  good nor bad, which is why recall quality sat at its starting value and no
+  source ever showed a settled quality signal however much the memory was used.
+- **The behavioural half of consolidation had never run.** Nothing triggered it,
+  and when it did run, two of its four steps failed without reporting anything.
+  SuperLocalMemory now runs it on a schedule while idle — never while you are
+  storing or recalling — and at session end.
+- **"This project" summaries could not work.** The button sent no project, and
+  the answer was always an error. SuperLocalMemory installs once for your whole
+  machine, so a browser tab has no way to know which project you mean; it now
+  lists the ones it has seen.
+- **Locally written summaries began with an apology.** The local model was never
+  told what it was writing, so it replied as if in a chat. It now returns the
+  summary and nothing else.
+- **A summary built from nothing claimed to cover everything.** A project with
+  recorded activity but no saved facts is now reported as a partial view.
+- **Multi-Agent Memory showed access tokens instead of agent names**, and because
+  one agent uses many tokens over time, a single agent appeared as dozens.
+- **Optimize invented a chart.** The cache hit-rate graph drew a rising trend
+  generated from the current value rather than measured; no history is recorded,
+  so it is gone until there is something real to plot. "Tokens saved" was
+  labelled as a monthly figure with an upward arrow that nothing ever set — it is
+  a total since install, and now says so. The page also explains why its numbers
+  are zero while its switches are on.
+- **One kind of memory pattern was never produced.** The query that finds it
+  asked for columns that do not exist, so it failed on every run and was skipped
+  silently.
+- **Code links never appeared for anyone.** Two separate reasons. The pass that
+  creates them matched no facts at all on a real store — it looked for an empty
+  archive marker, while live facts carry `live`, so every fact was excluded. And
+  the panel that displayed them was the search-results view, not the memory
+  drawer people actually open. Both fixed; on a real store this produces 7,470
+  links across 3,603 memories.
+- **"Atomic facts" always read "No atomic facts recorded for this memory."** The
+  drawer looked up child facts using the row's own id, but a row in that pane is
+  itself an atomic fact and its parent is a separate field. Records stored
+  directly, which have no parent, now say so rather than reporting missing data.
+- **Summaries ignored your mode.** The command never passed the configuration, so
+  every summary took the no-model path — a Mode B user got Mode A output. Ollama
+  now writes them when configured, falling back if it is unavailable.
+- **Summary highlights were unreadable.** Truncation counted characters but
+  ignored newlines, so one fact containing headings and blank lines rendered as
+  six lines and the list stopped being a list.
+- **"Active entities" listed internal identifiers** instead of names. It now
+  reads `Fixed (127), Gateway (124)`.
+
 ## [4.0.7] — Reachable
 
 Everything here was already built. None of it could be used.
@@ -871,7 +1089,7 @@ Eight community pull requests merged after line-by-line review, plus fixes for t
 
 ## [3.6.14] - 2026-06-18 — Audit-hardened: memory bounds, cross-tenant cache isolation, atomic credentials
 
-Shipped through two adversarial audit passes (Qualixar Iron Pattern Stages 8–9), validated against a green 5933-test suite under the real 3.12 runtime. Default single-machine behavior is unchanged.
+Default single-machine behavior is unchanged.
 
 ### Security
 

package/README.md

@@ -5,7 +5,7 @@
   </picture>
 </p>
 
-<h1 align="center">SuperLocalMemory V4.0.7</h1>
+<h1 align="center">SuperLocalMemory V4.0.9</h1>
 
 <h2 align="center">Rent the LLM. Own the memory.</h2>
 
@@ -27,12 +27,12 @@ guarantee here is stated as a falsifiable invariant, tested under an adversarial
 negative control, and shipped with the harness that regenerates the evidence:
 <code>python benchmark/run_all.py --trials 200 --output-dir results/</code>. What each experiment
 does <em>not</em> exercise is stated too.</p>
-<p align="center"><code>v4.0.7</code> — one control plane: <strong>SLM-Mesh</strong> peer coordination · multi-scope memory (personal / shared / global) · profiles · Cache · Compress · 7-layer retrieval · code graph · Entity Explorer · skill evolution · Modes A/B/C · GDPR retention &amp; audit chain · bounded loops — across CLI, MCP, dashboard, the <strong>Claude plugin</strong>, the <strong>Codex add-on</strong>, and documented IDE integrations.<br/>
+<p align="center"><code>v4.0.9</code> — one control plane: <strong>SLM-Mesh</strong> peer coordination · multi-scope memory (personal / shared / global) · profiles · Cache · Compress · 7-layer retrieval · code graph · Entity Explorer · skill evolution · Modes A/B/C · GDPR retention &amp; audit chain · bounded loops — across CLI, MCP, dashboard, the <strong>Claude plugin</strong>, the <strong>Codex add-on</strong>, and documented IDE integrations.<br/>
 Proxy: <code>slm wrap claude</code> &nbsp;·&nbsp; MCP: add <code>slm_compress</code> to your config &nbsp;·&nbsp; Skill: zero-config</p>
 <p align="center"><strong>Four public arXiv preprints</strong> · V4: <a href="https://arxiv.org/abs/2608.08253">arXiv:2608.08253</a> · companion archive: <a href="https://zenodo.org/records/21853302">Zenodo 21853302</a> (<a href="https://doi.org/10.5281/zenodo.21853302">DOI 10.5281/zenodo.21853302</a>) · prior preprints: <a href="https://arxiv.org/abs/2603.02240">2603.02240</a> · <a href="https://arxiv.org/abs/2603.14588">2603.14588</a> · <a href="https://arxiv.org/abs/2604.04514">2604.04514</a>.</p>
 
 <p align="center">
-  <a href="CHANGELOG.md"><img src="https://img.shields.io/badge/v4.0.7-Current_Release-2ea44f?style=for-the-badge&logo=checkmarx&logoColor=white" alt="v4.0.7 — Current Release"/></a>
+  <a href="CHANGELOG.md"><img src="https://img.shields.io/badge/v4.0.9-Current_Release-2ea44f?style=for-the-badge&logo=checkmarx&logoColor=white" alt="v4.0.9 — Current Release"/></a>
   <a href="https://arxiv.org/abs/2608.08253"><img src="https://img.shields.io/badge/arXiv-2608.08253-b31b1b?style=for-the-badge&logo=arxiv&logoColor=white" alt="SuperLocalMemory 4.0 paper on arXiv:2608.08253"/></a>
   <a href="https://zenodo.org/records/21853302"><img src="https://img.shields.io/badge/Zenodo-10.5281%2Fzenodo.21853302-1682D4?style=for-the-badge&logo=zenodo&logoColor=white" alt="V4 paper on Zenodo: 10.5281/zenodo.21853302"/></a>
   <a href="https://arxiv.org/abs/2603.14588"><img src="https://img.shields.io/badge/arXiv-2603.14588-b31b1b?style=for-the-badge&logo=arxiv&logoColor=white" alt="arXiv Paper"/></a>
@@ -386,8 +386,8 @@ Full docs: [docs/multi-machine.md](docs/multi-machine.md) · [docs/distributed-d
 | **Python CLI + SDK** (primary) | Activate a Python virtual environment, then `python -m pip install superlocalmemory` | Python 3.11+; the `slm` CLI and importable SDK stay inside that environment |
 | **Repository clone — macOS/Linux** | `./scripts/install.sh install` | Research/contributor path; delegates to an existing uv or pipx installation |
 | **Repository clone — Windows** | `.\scripts\install.ps1 -Action Install` | Research/contributor path; delegates to an existing uv or pipx installation |
-| **Claude Code Plugin** (WP-06) | `/plugin marketplace add qualixar/superlocalmemory` then `/plugin install superlocalmemory@qualixar` | Self-bootstraps venv, isolated SLM_DATA_DIR, additive — 16-tool core. Ships the skills/agents/hooks/commands |
-| **Portable / IDE connect** (WP-08) | `slm connect <ide> [--here]` | Wire any IDE without reinstalling; `slm connect claude-code` → plugin pointer |
+| **Claude Code Plugin** | `/plugin marketplace add qualixar/superlocalmemory` then `/plugin install superlocalmemory@qualixar` | Self-bootstraps venv, isolated SLM_DATA_DIR, additive — 16-tool core. Ships the skills/agents/hooks/commands |
+| **Portable / IDE connect** | `slm connect <ide> [--here]` | Wire any IDE without reinstalling; `slm connect claude-code` → plugin pointer |
 
 After any install path: `slm setup` → `slm doctor` → `slm warmup` (optional, pre-downloads ~500MB embedding model).
 
@@ -433,7 +433,7 @@ Or: `claude mcp add --transport http superlocalmemory http://127.0.0.1:8765/mcp/
 { "mcpServers": { "superlocalmemory": { "command": "slm", "args": ["mcp"] } } }
 ```
 
-### MCP Profiles (WP-01)
+### MCP Profiles
 
 Control tool surface via `SLM_MCP_PROFILE`:
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "superlocalmemory",
-  "version": "4.0.7",
+  "version": "4.0.9",
   "description": "Local-first agent memory with MCP and an agent-native CLI. Documented clients include Claude Code, Cursor, and Windsurf.",
   "keywords": [
     "ai-memory",

package/plugin/.claude-plugin/plugin.json

@@ -15,5 +15,5 @@
   "mcpServers": "./.mcp.json",
   "name": "superlocalmemory",
   "repository": "https://github.com/qualixar/superlocalmemory",
-  "version": "4.0.7"
+  "version": "4.0.9"
 }

package/plugin/CLAUDE.md

@@ -1,4 +1,4 @@
-<!-- BEGIN SuperLocalMemory v4.0.7 -->
+<!-- BEGIN SuperLocalMemory v4.0.9 -->
 
 ## SuperLocalMemory (SLM) — Agent Rules
 
@@ -39,6 +39,6 @@ slm-recall · slm-remember · slm-session · slm-status · slm-cache · slm-comp
 ### Subagents
 slm-memory-advisor (memory decisions, session hygiene, scope/profile guidance) · slm-optimize-advisor (context compression + KV cache) · slm-governance-advisor (scope/roles/compliance/GDPR)
 
-<!-- END SuperLocalMemory v4.0.7 -->
+<!-- END SuperLocalMemory v4.0.9 -->
 
-SuperLocalMemory v4.0.7 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.9 · Qualixar · AGPL-3.0-or-later

package/plugin/agents/slm-governance-advisor.md

@@ -77,4 +77,4 @@ slm-scope · slm-governance · slm-profile · slm-remember · slm-recall
 # What NOT to do
 Never session_init twice; never forget without dry-run preview; never store secrets; never bypass role checks; never claim an erasure succeeded without verifying via recall.
 
-SuperLocalMemory v4.0.7 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.9 · Qualixar · AGPL-3.0-or-later

package/plugin/agents/slm-loop-runner.md

@@ -68,4 +68,4 @@ assessment. The gate is the authority.
 
 ---
 
-SuperLocalMemory v4.0.7 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.9 · Qualixar · AGPL-3.0-or-later

package/plugin/agents/slm-memory-advisor.md

@@ -46,4 +46,4 @@ slm-recall · slm-remember · slm-session · slm-scope · slm-profile · slm-gov
 # What NOT to do
 Never session_init twice; never forget dry_run=False without reporting preview; never dump a whole file into remember; never invent a memory; never claim "saved" without success:true / clean CLI exit; never bypass scope or governance restrictions.
 
-SuperLocalMemory v4.0.7 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.9 · Qualixar · AGPL-3.0-or-later

package/plugin/agents/slm-optimize-advisor.md

@@ -41,4 +41,4 @@ slm-compress · slm-cache · slm-status · slm-profile
 # What NOT to do
 Never compress code-for-edit/JSON-to-parse/<500 chars; never store secrets/ccr_ids; never let optimize failure block/alter the task; never claim a specific savings %; never carry ccr_ids across profile switches.
 
-SuperLocalMemory v4.0.7 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.9 · Qualixar · AGPL-3.0-or-later

package/plugin/requirements.txt

@@ -1 +1 @@
-superlocalmemory==4.0.7
+superlocalmemory==4.0.9

package/plugin/scripts/ensure-venv.sh

@@ -1,5 +1,5 @@
 #!/usr/bin/env bash
-# ensure-venv.sh — WP-06 SuperLocalMemory plugin venv bootstrap
+# ensure-venv.sh — SuperLocalMemory plugin venv bootstrap
 #
 # Called by Claude Code SessionStart hook. Idempotent: fast-path exits in <100ms
 # on repeat invocations when requirements.txt is unchanged.

package/plugin/skills/slm-cache/SKILL.md

@@ -145,4 +145,4 @@ These subcommands control daemon-level cache settings. They do not read or write
 
 ---
 
-SuperLocalMemory v4.0.7 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.9 · Qualixar · AGPL-3.0-or-later

package/plugin/skills/slm-compress/SKILL.md

@@ -147,4 +147,4 @@ Content over 1 MB (1 000 000 bytes UTF-8) is processed but `reversible` is force
 
 ---
 
-SuperLocalMemory v4.0.7 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.9 · Qualixar · AGPL-3.0-or-later

package/plugin/skills/slm-governance/SKILL.md

@@ -245,4 +245,4 @@ Before running any destructive operation (`forget`, `compact_memories`):
 
 ---
 
-*SuperLocalMemory v4.0.7 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.9 · Qualixar · AGPL-3.0-or-later*

package/plugin/skills/slm-graph/SKILL.md

@@ -312,4 +312,4 @@ profile. See `slm-profile` for the full profile switching workflow.
 
 ---
 
-SuperLocalMemory v4.0.7 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.9 · Qualixar · AGPL-3.0-or-later

package/plugin/skills/slm-loop/SKILL.md

@@ -96,4 +96,4 @@ paused, name the approval needed; when errored, quote the short detail.
 
 ---
 
-SuperLocalMemory v4.0.7 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.9 · Qualixar · AGPL-3.0-or-later

package/plugin/skills/slm-mesh/SKILL.md

@@ -279,4 +279,4 @@ mesh availability.
 
 ---
 
-*SuperLocalMemory v4.0.7 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.9 · Qualixar · AGPL-3.0-or-later*

package/plugin/skills/slm-profile/SKILL.md

@@ -146,4 +146,4 @@ Name them differently in your MCP config (e.g. `superlocalmemory-personal` and
 
 ---
 
-*SuperLocalMemory v4.0.7 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.9 · Qualixar · AGPL-3.0-or-later*

package/plugin/skills/slm-recall/SKILL.md

@@ -236,4 +236,4 @@ before recalling, then switch back. See `slm-profile` for workspace switching.
 
 ---
 
-*SuperLocalMemory v4.0.7 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.9 · Qualixar · AGPL-3.0-or-later*

package/plugin/skills/slm-remember/SKILL.md

@@ -238,4 +238,4 @@ different workspace, use `switch_profile` first. See `slm-profile`.
 
 ---
 
-*SuperLocalMemory v4.0.7 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.9 · Qualixar · AGPL-3.0-or-later*

package/plugin/skills/slm-scope/SKILL.md

@@ -173,4 +173,4 @@ to review the impact. See `slm-remember` for the full deletion discipline.
 
 ---
 
-*SuperLocalMemory v4.0.7 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.9 · Qualixar · AGPL-3.0-or-later*

package/plugin/skills/slm-session/SKILL.md

@@ -227,4 +227,4 @@ explicitly and call `recall` with `include_global`/`include_shared` after
 
 ---
 
-*SuperLocalMemory v4.0.7 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.9 · Qualixar · AGPL-3.0-or-later*

package/plugin/skills/slm-status/SKILL.md

@@ -61,7 +61,7 @@ If `cache_kv_hits` is 0 after repeated work, verify key naming consistency (the
 slm status [--json] [--verbose]
 ```
 
-Reports system-level state — not optimization counters. Canonical fields (WP-02):
+Reports system-level state — not optimization counters. Canonical fields:
 
 - **mode** — active operation mode (e.g. `local`)
 - **profile** — current memory profile name
@@ -101,7 +101,7 @@ Do not rely on the human-readable format for parsing — always use `--json` whe
 slm doctor [--json] [--quick]
 ```
 
-Preflight check covering dependencies, embedding worker, daemon connectivity, and Surface B health. The **"Optimize (Surface B)"** line (WP-03) confirms whether the compression and cache subsystem initialised correctly.
+Preflight check covering dependencies, embedding worker, daemon connectivity, and Surface B health. The **"Optimize (Surface B)"** line confirms whether the compression and cache subsystem initialised correctly.
 
 `--quick` skips the daemon and embedding probes — runs only dependency and config checks; faster but incomplete.
 
@@ -163,4 +163,4 @@ multi-profile setup. To switch the active profile, see `slm-profile`.
 
 ---
 
-SuperLocalMemory v4.0.7 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.9 · Qualixar · AGPL-3.0-or-later

package/plugin-src/rules/AGENTS.md

@@ -129,4 +129,4 @@ When the SLM MCP server is unavailable, use these CLI equivalents:
 - **slm-optimize-advisor** — context compression and KV cache
 - **slm-governance-advisor** — scope/role compliance, retention policies, GDPR
 
-SuperLocalMemory v4.0.7 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.9 · Qualixar · AGPL-3.0-or-later

package/plugin-src/skills/slm-status/SKILL.md

@@ -61,7 +61,7 @@ If `cache_kv_hits` is 0 after repeated work, verify key naming consistency (the
 slm status [--json] [--verbose]
 ```
 
-Reports system-level state — not optimization counters. Canonical fields (WP-02):
+Reports system-level state — not optimization counters. Canonical fields:
 
 - **mode** — active operation mode (e.g. `local`)
 - **profile** — current memory profile name
@@ -101,7 +101,7 @@ Do not rely on the human-readable format for parsing — always use `--json` whe
 slm doctor [--json] [--quick]
 ```
 
-Preflight check covering dependencies, embedding worker, daemon connectivity, and Surface B health. The **"Optimize (Surface B)"** line (WP-03) confirms whether the compression and cache subsystem initialised correctly.
+Preflight check covering dependencies, embedding worker, daemon connectivity, and Surface B health. The **"Optimize (Surface B)"** line confirms whether the compression and cache subsystem initialised correctly.
 
 `--quick` skips the daemon and embedding probes — runs only dependency and config checks; faster but incomplete.
 

package/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "superlocalmemory"
-version = "4.0.7"
+version = "4.0.9"
 description = "Local-first agent memory with auditable hybrid retrieval"
 readme = "README.md"
 license = "AGPL-3.0-or-later"

package/scripts/postinstall.js

@@ -201,6 +201,10 @@ function main(argv = process.argv.slice(2)) {
   console.log(`SuperLocalMemory ${packageVersion}: isolated runtime verified.`);
   console.log('No memory database, IDE hooks, daemon, configuration, or models were changed.');
   console.log('');
+  console.log('  Your database will be automatically migrated on first run.');
+  console.log('  A backup is created in ~/.superlocalmemory/pre-migration-snapshots/ before');
+  console.log('  any migration runs. No action is needed from you.');
+  console.log('');
   console.log('  Next step — run the guided setup (picks your mode, downloads');
   console.log('  models, connects your IDEs). Takes about a minute:');
   console.log('');

package/src/superlocalmemory/__init__.py

@@ -32,7 +32,7 @@ if "OMP_NUM_THREADS" not in os.environ:
     os.environ["OMP_NUM_THREADS"] = "2"
 # ---------------------------------------------------------------------------
 
-__version__ = "4.0.7"
+__version__ = "4.0.9"
 
 _REQUIRED_VERSIONS = {
     "sentence_transformers": "5.3.0",

package/src/superlocalmemory/cli/_lazy_init.py

@@ -2,7 +2,7 @@
 # Licensed under AGPL-3.0-or-later - see LICENSE file
 # Part of SuperLocalMemory V3 | https://qualixar.com | https://varunpratap.com
 
-"""WP-07 — Lazy first-run initialisation (pip cross-install).
+"""Lazy first-run initialisation for pip cross-installs.
 
 Lightweight import only — zero heavy imports, STDOUT-SILENT.