superlocalmemory 4.1.14 → 4.1.15
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/.claude-plugin/marketplace.json +2 -2
- package/CHANGELOG.md +55 -0
- package/README.md +3 -3
- package/package.json +1 -1
- package/plugin/.claude-plugin/plugin.json +1 -1
- package/plugin/CLAUDE.md +3 -3
- package/plugin/agents/slm-governance-advisor.md +1 -1
- package/plugin/agents/slm-loop-runner.md +1 -1
- package/plugin/agents/slm-memory-advisor.md +1 -1
- package/plugin/agents/slm-optimize-advisor.md +1 -1
- package/plugin/requirements.txt +1 -1
- package/plugin/skills/slm-cache/SKILL.md +1 -1
- package/plugin/skills/slm-compress/SKILL.md +1 -1
- package/plugin/skills/slm-governance/SKILL.md +1 -1
- package/plugin/skills/slm-graph/SKILL.md +1 -1
- package/plugin/skills/slm-loop/SKILL.md +1 -1
- package/plugin/skills/slm-mesh/SKILL.md +1 -1
- package/plugin/skills/slm-profile/SKILL.md +1 -1
- package/plugin/skills/slm-recall/SKILL.md +1 -1
- package/plugin/skills/slm-remember/SKILL.md +1 -1
- package/plugin/skills/slm-scope/SKILL.md +1 -1
- package/plugin/skills/slm-session/SKILL.md +1 -1
- package/plugin/skills/slm-status/SKILL.md +1 -1
- package/plugin-src/agents/slm-memory-advisor.md +1 -1
- package/plugin-src/agents/slm-optimize-advisor.md +1 -1
- package/plugin-src/rules/AGENTS.md +1 -1
- package/plugin-src/skills/slm-cache/SKILL.md +1 -1
- package/plugin-src/skills/slm-compress/SKILL.md +1 -1
- package/plugin-src/skills/slm-governance/SKILL.md +1 -1
- package/plugin-src/skills/slm-graph/SKILL.md +1 -1
- package/plugin-src/skills/slm-loop/SKILL.md +1 -1
- package/plugin-src/skills/slm-mesh/SKILL.md +1 -1
- package/plugin-src/skills/slm-profile/SKILL.md +1 -1
- package/plugin-src/skills/slm-recall/SKILL.md +1 -1
- package/plugin-src/skills/slm-remember/SKILL.md +1 -1
- package/plugin-src/skills/slm-scope/SKILL.md +1 -1
- package/plugin-src/skills/slm-session/SKILL.md +1 -1
- package/plugin-src/skills/slm-status/SKILL.md +1 -1
|
@@ -21,7 +21,7 @@
|
|
|
21
21
|
"license": "AGPL-3.0-or-later",
|
|
22
22
|
"name": "superlocalmemory",
|
|
23
23
|
"source": "./plugin",
|
|
24
|
-
"version": "4.1.
|
|
24
|
+
"version": "4.1.15"
|
|
25
25
|
},
|
|
26
26
|
{
|
|
27
27
|
"author": {
|
|
@@ -39,7 +39,7 @@
|
|
|
39
39
|
"license": "AGPL-3.0-or-later",
|
|
40
40
|
"name": "superlocalmemory-codex",
|
|
41
41
|
"source": "./codex-plugin",
|
|
42
|
-
"version": "4.1.
|
|
42
|
+
"version": "4.1.15"
|
|
43
43
|
}
|
|
44
44
|
]
|
|
45
45
|
}
|
package/CHANGELOG.md
CHANGED
|
@@ -5,6 +5,61 @@ All notable changes to SuperLocalMemory will be documented in this file.
|
|
|
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).
|
|
7
7
|
|
|
8
|
+
## [4.1.15] — Retrieval-honesty release
|
|
9
|
+
|
|
10
|
+
### Fixed
|
|
11
|
+
|
|
12
|
+
- **A corrected memory stops being served as current.** Correcting a memory
|
|
13
|
+
made search worse than leaving the wrong one in place: the retirement was
|
|
14
|
+
recorded and recall honoured it, but every other read path kept serving the
|
|
15
|
+
old memory at full weight, and no amount of waiting changed that. On a real
|
|
16
|
+
5,560-memory store, 4.8% of search results were memories the store had
|
|
17
|
+
already decided were wrong — up to 12% on a topic-specific search. Search,
|
|
18
|
+
pinned context, the degraded start-up path and the scheduled-memory surface
|
|
19
|
+
now all resolve visibility from one place. (#136)
|
|
20
|
+
- **The degraded start-up path filtered nothing.** When the daemon is
|
|
21
|
+
unreachable, session start-up falls back to a direct search. That fallback
|
|
22
|
+
applied no visibility rules at all, so it could hand an agent a withheld or
|
|
23
|
+
retired memory as opening context.
|
|
24
|
+
- **`fetch` says which ids it could not find.** It returned success with an
|
|
25
|
+
empty result for an id it could not read — indistinguishable from the
|
|
26
|
+
answer for a memory that does not exist, on the one call an agent uses to
|
|
27
|
+
check that a write landed. It now names what it could not resolve, reports
|
|
28
|
+
partial misses, and accepts a list as well as a comma-separated string.
|
|
29
|
+
(#135)
|
|
30
|
+
- **A new memory starts active instead of landing anywhere.** Where a memory
|
|
31
|
+
sat in its lifecycle was decided by a random simulation rather than by how
|
|
32
|
+
it had been used. Memories saved seconds apart could land in three
|
|
33
|
+
different tiers, and a freshly written memory could never reach the
|
|
34
|
+
most-available tier at all — on one store, 2 memories out of 5,560 were
|
|
35
|
+
active and 93.6% were marked archived, so features that read only active
|
|
36
|
+
memories were reading almost nothing. A memory now starts fully available
|
|
37
|
+
and moves down only as it goes unused, recovering when used again. The same
|
|
38
|
+
memory always lands in the same place. Existing stores are recomputed on
|
|
39
|
+
the next maintenance pass. (#136)
|
|
40
|
+
- **Two causes of wasted background work.** Reconfiguring or switching profile
|
|
41
|
+
built a new set of storage backends and abandoned the old ones without
|
|
42
|
+
closing them, leaving a native thread pool and an orphaned background worker
|
|
43
|
+
running for the life of the process. Two background loops also treated a
|
|
44
|
+
failure as a reason to retry immediately, holding them at full speed on work
|
|
45
|
+
that could never succeed. Both are fixed.
|
|
46
|
+
|
|
47
|
+
**This does not close #137.** The dominant cause of the reported high CPU is
|
|
48
|
+
separate and still open: the vector projection is written one memory at a
|
|
49
|
+
time, each write creates a new version of the vector store, and nothing ever
|
|
50
|
+
removes old versions. Measured on the author's machine: 5,561 memories, a
|
|
51
|
+
610 MB database — and a 17 GB vector store holding 50,580 versions. Every
|
|
52
|
+
vector operation walks that history, which is why CPU stays pinned with no
|
|
53
|
+
work queued. Setting `vector_backend` to `sqlite-vec` avoids it today.
|
|
54
|
+
|
|
55
|
+
### Added
|
|
56
|
+
|
|
57
|
+
- **Ask a busy daemon what it is doing.** `kill -USR1 <pid>` writes every
|
|
58
|
+
thread's stack to `thread-dump.log`. Diagnosing the high-CPU report needed
|
|
59
|
+
root on macOS and defeated the standard profiler on Linux, so on both
|
|
60
|
+
machines nobody could see inside the process. Python frames only — a native
|
|
61
|
+
thread with no Python frame stays invisible.
|
|
62
|
+
|
|
8
63
|
## [4.1.14] — Profile-correctness and reliability release
|
|
9
64
|
|
|
10
65
|
### Added
|
package/README.md
CHANGED
|
@@ -5,7 +5,7 @@
|
|
|
5
5
|
</picture>
|
|
6
6
|
</p>
|
|
7
7
|
|
|
8
|
-
<h1 align="center">SuperLocalMemory V4.1.
|
|
8
|
+
<h1 align="center">SuperLocalMemory V4.1.15</h1>
|
|
9
9
|
|
|
10
10
|
<h2 align="center">Rent the LLM. Own the memory.</h2>
|
|
11
11
|
|
|
@@ -27,12 +27,12 @@ guarantee here is stated as a falsifiable invariant, tested under an adversarial
|
|
|
27
27
|
negative control, and shipped with the harness that regenerates the evidence:
|
|
28
28
|
<code>python benchmark/run_all.py --trials 200 --output-dir results/</code>. What each experiment
|
|
29
29
|
does <em>not</em> exercise is stated too.</p>
|
|
30
|
-
<p align="center"><code>v4.1.
|
|
30
|
+
<p align="center"><code>v4.1.15</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 & audit chain · bounded loops — across CLI, MCP, dashboard, the <strong>Claude plugin</strong>, the <strong>Codex add-on</strong>, and documented IDE integrations.<br/>
|
|
31
31
|
Proxy: <code>slm wrap claude</code> · MCP: add <code>slm_compress</code> to your config · Skill: zero-config</p>
|
|
32
32
|
<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>
|
|
33
33
|
|
|
34
34
|
<p align="center">
|
|
35
|
-
<a href="CHANGELOG.md"><img src="https://img.shields.io/badge/v4.1.
|
|
35
|
+
<a href="CHANGELOG.md"><img src="https://img.shields.io/badge/v4.1.15-Current_Release-2ea44f?style=for-the-badge&logo=checkmarx&logoColor=white" alt="v4.1.15 — Current Release"/></a>
|
|
36
36
|
<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>
|
|
37
37
|
<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>
|
|
38
38
|
<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>
|
package/package.json
CHANGED
package/plugin/CLAUDE.md
CHANGED
|
@@ -1,4 +1,4 @@
|
|
|
1
|
-
<!-- BEGIN SuperLocalMemory v4.1.
|
|
1
|
+
<!-- BEGIN SuperLocalMemory v4.1.15 -->
|
|
2
2
|
|
|
3
3
|
## SuperLocalMemory (SLM) — Agent Rules
|
|
4
4
|
|
|
@@ -39,6 +39,6 @@ slm-recall · slm-remember · slm-session · slm-status · slm-cache · slm-comp
|
|
|
39
39
|
### Subagents
|
|
40
40
|
slm-memory-advisor (memory decisions, session hygiene, scope/profile guidance) · slm-optimize-advisor (context compression + KV cache) · slm-governance-advisor (scope/roles/compliance/GDPR)
|
|
41
41
|
|
|
42
|
-
<!-- END SuperLocalMemory v4.1.
|
|
42
|
+
<!-- END SuperLocalMemory v4.1.15 -->
|
|
43
43
|
|
|
44
|
-
SuperLocalMemory v4.1.
|
|
44
|
+
SuperLocalMemory v4.1.15 · Qualixar · AGPL-3.0-or-later
|
|
@@ -77,4 +77,4 @@ slm-scope · slm-governance · slm-profile · slm-remember · slm-recall
|
|
|
77
77
|
# What NOT to do
|
|
78
78
|
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.
|
|
79
79
|
|
|
80
|
-
SuperLocalMemory v4.1.
|
|
80
|
+
SuperLocalMemory v4.1.15 · Qualixar · AGPL-3.0-or-later
|
|
@@ -46,4 +46,4 @@ slm-recall · slm-remember · slm-session · slm-scope · slm-profile · slm-gov
|
|
|
46
46
|
# What NOT to do
|
|
47
47
|
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.
|
|
48
48
|
|
|
49
|
-
SuperLocalMemory v4.1.
|
|
49
|
+
SuperLocalMemory v4.1.15 · Qualixar · AGPL-3.0-or-later
|
|
@@ -41,4 +41,4 @@ slm-compress · slm-cache · slm-status · slm-profile
|
|
|
41
41
|
# What NOT to do
|
|
42
42
|
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.
|
|
43
43
|
|
|
44
|
-
SuperLocalMemory v4.1.
|
|
44
|
+
SuperLocalMemory v4.1.15 · Qualixar · AGPL-3.0-or-later
|
package/plugin/requirements.txt
CHANGED
|
@@ -1 +1 @@
|
|
|
1
|
-
superlocalmemory==4.1.
|
|
1
|
+
superlocalmemory==4.1.15
|
|
@@ -46,4 +46,4 @@ slm-recall · slm-remember · slm-session · slm-scope · slm-profile · slm-gov
|
|
|
46
46
|
# What NOT to do
|
|
47
47
|
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.
|
|
48
48
|
|
|
49
|
-
SuperLocalMemory v4.1.
|
|
49
|
+
SuperLocalMemory v4.1.15 · Qualixar · AGPL-3.0-or-later
|
|
@@ -41,4 +41,4 @@ slm-compress · slm-cache · slm-status · slm-profile
|
|
|
41
41
|
# What NOT to do
|
|
42
42
|
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.
|
|
43
43
|
|
|
44
|
-
SuperLocalMemory v4.1.
|
|
44
|
+
SuperLocalMemory v4.1.15 · Qualixar · AGPL-3.0-or-later
|
|
@@ -137,4 +137,4 @@ When the SLM MCP server is unavailable, use these CLI equivalents:
|
|
|
137
137
|
- **slm-optimize-advisor** — context compression and KV cache
|
|
138
138
|
- **slm-governance-advisor** — scope/role compliance, retention policies, GDPR
|
|
139
139
|
|
|
140
|
-
SuperLocalMemory v4.1.
|
|
140
|
+
SuperLocalMemory v4.1.15 · Qualixar · AGPL-3.0-or-later
|