superlocalmemory 3.4.21 → 3.4.23

package/CHANGELOG.md

@@ -10,7 +10,36 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ---
 
-## [3.4.21] - 2026-04-18
+## [3.4.23] - 2026-04-21
+
+Critical hotfix on top of 3.4.22 for two end-user-facing regressions.
+
+### Fixed
+- **Daemon error log no longer balloons.** A ternary passed as the
+  `logger.info` format string caused a `TypeError` on every startup in 24/7
+  mode. Python's logging module then dumped the full FastAPI
+  `merged_lifespan` stack to stderr; over a day the LaunchAgent log grew to
+  tens of MB. The call is now pre-formatted. A defensive log-rotation pass
+  at startup truncates any daemon log over 10 MB so users upgrading from
+  3.4.22 get a clean slate on first boot.
+- **Dashboard no longer hangs after a daemon upgrade.** Static JS/CSS/HTML
+  was served without cache headers, so browsers served stale modules after
+  `slm restart` and the dashboard showed an infinite spinner. All static
+  responses now ship `Cache-Control: no-cache, must-revalidate`, and
+  `index.html` embeds the server version; on mismatch the tab clears
+  `localStorage` (preserving theme) and hard-reloads once.
+- **Fetches can no longer hang forever.** A global `fetch` patch attaches a
+  15-second `AbortController` timeout to every relative-URL request, so a
+  dead socket surfaces as a rejection instead of leaving a spinner
+  spinning. No callsite changes required.
+
+### Added
+- `GET /api/version` — returns the running daemon version; consumed by the
+  dashboard version-fingerprint auto-reload.
+
+---
+
+## [3.4.22] - 2026-04-18
 
 Hardening release — correctness, stability, and security fixes.
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "superlocalmemory",
-  "version": "3.4.21",
+  "version": "3.4.23",
   "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.",
   "keywords": [
     "ai-memory",

package/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "superlocalmemory"
-version = "3.4.21"
+version = "3.4.23"
 description = "Information-geometric agent memory with mathematical guarantees"
 readme = "README.md"
 license = {text = "AGPL-3.0-or-later"}
@@ -116,7 +116,7 @@ build-backend = "setuptools.build_meta"
 where = ["src"]
 
 [tool.setuptools.package-data]
-superlocalmemory = ["ui/**/*"]
+superlocalmemory = ["ui/**/*", "skills/**/*"]
 
 [tool.pytest.ini_options]
 testpaths = ["tests"]

package/scripts/build_entry.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-06 §3
+# Part of SuperLocalMemory v3.4.22 — LLD-06 §3
 
 """AST-extract binary hook entry.
 

package/scripts/release_manifest.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-06 §6.1
+# Part of SuperLocalMemory v3.4.22 — LLD-06 §6.1
 
 """Generate the release ``manifest.json`` for ``slm-hook`` binaries.
 
@@ -10,7 +10,7 @@ Section 6.1 (Release manifest).
 Shape (per LLD-06 §6.1):
 
     {
-      "version": "3.4.21",
+      "version": "3.4.22",
       "released_at": "2026-04-17T00:00:00Z",
       "assets": [
         {

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

@@ -1,7 +1,7 @@
 ---
 name: slm-build-graph
 description: Build or rebuild the knowledge graph from existing memories using TF-IDF entity extraction and Leiden clustering. Use when search results seem poor, after bulk imports, or to optimize performance. Automatically discovers relationships between memories and creates topic clusters.
-version: "3.4.21"
+version: "3.4.23"
 license: AGPL-3.0-or-later
 compatibility: "Requires SuperLocalMemory V2 installed at ~/.claude-memory/, optional dependencies: python-igraph, leidenalg"
 attribution:

package/skills/slm-list-recent/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: slm-list-recent
 description: List most recent memories in chronological order. Use when the user wants to see what was recently saved, review recent conversations, check what they worked on today, or browse memory history. Shows memories sorted by creation time (newest first).
-version: "3.4.21"
+version: "3.4.23"
 license: AGPL-3.0-or-later
 compatibility: "Requires SuperLocalMemory V2 installed at ~/.claude-memory/"
 attribution:

package/skills/slm-recall/SKILL.md

@@ -1,9 +1,9 @@
 ---
 name: slm-recall
 description: Search SuperLocalMemory for relevant facts, decisions, and past context. Use when the user asks to recall, search, find, or retrieve stored information. Invokes 5-channel retrieval with LightGBM reranking via MCP.
-version: "3.4.21"
+version: "3.4.23"
 license: AGPL-3.0-or-later
-compatibility: "SuperLocalMemory v3.4.21 — MCP (preferred) or CLI fallback"
+compatibility: "SuperLocalMemory v3.4.22 — MCP (preferred) or CLI fallback"
 attribution:
   creator: Varun Pratap Bhardwaj
   role: Solution Architect & Creator
@@ -325,7 +325,7 @@ slm recall "query" --format csv
 
 *Open source doesn't mean removing credit. Attribution must be preserved per MIT License terms.*
 
-## V3 Update (v3.4.21)
+## V3 Update (v3.4.22)
 
 **Preferred**: Use the `recall` MCP tool directly — it uses 5-channel retrieval
 (semantic, BM25, entity-graph, temporal, Hopfield) with LightGBM reranking

package/skills/slm-remember/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: slm-remember
 description: Save content to SuperLocalMemory with intelligent indexing and knowledge graph integration. Use when the user wants to remember information, save context, store coding decisions, or persist knowledge for future sessions. Automatically indexes, graphs, and learns patterns.
-version: "3.4.21"
+version: "3.4.23"
 license: AGPL-3.0-or-later
 compatibility: "Requires SuperLocalMemory V2 installed at ~/.claude-memory/"
 attribution:

package/skills/slm-status/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: slm-status
 description: Check SuperLocalMemory system status, health, and statistics. Use when the user wants to know memory count, graph stats, patterns learned, database health, or system diagnostics. Shows comprehensive system health dashboard.
-version: "3.4.21"
+version: "3.4.23"
 license: AGPL-3.0-or-later
 compatibility: "Requires SuperLocalMemory V2 installed at ~/.claude-memory/"
 attribution:

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

@@ -1,7 +1,7 @@
 ---
 name: slm-switch-profile
 description: Switch between memory profiles for context isolation and management. Use when the user wants to change profile context, separate work/personal memories, or manage multiple independent memory spaces. Each profile has its own database, graph, and patterns.
-version: "3.4.21"
+version: "3.4.23"
 license: AGPL-3.0-or-later
 compatibility: "Requires SuperLocalMemory V2 installed at ~/.claude-memory/"
 attribution:

package/src/superlocalmemory/__init__.py

@@ -0,0 +1,3 @@
+"""SuperLocalMemory — information-geometric agent memory."""
+
+__version__ = "3.4.23"

package/src/superlocalmemory/cli/commands.py

@@ -112,11 +112,11 @@ def dispatch(args: Namespace) -> None:
         # V3.4.11 skill evolution
         "config": cmd_config,
         "evolve": cmd_evolve,
-        # V3.4.21 LLD-05 context pre-staging
+        # V3.4.22 LLD-05 context pre-staging
         "context": _cmd_context_dispatch,
-        # V3.4.21 LLD-06 additive schema migrations
+        # V3.4.22 LLD-06 additive schema migrations
         "db": _cmd_db_dispatch,
-        # V3.4.21 Stage 8 SB-5 — MASTER-PLAN §8 escape hatches.
+        # V3.4.22 Stage 8 SB-5 — MASTER-PLAN §8 escape hatches.
         "disable": _cmd_escape_disable,
         "enable": _cmd_escape_enable,
         "clear-cache": _cmd_escape_clear_cache,
@@ -690,13 +690,13 @@ def cmd_provider(args: Namespace) -> None:
 
 
 def _cmd_context_dispatch(args: Namespace) -> None:
-    """V3.4.21 LLD-05: ``slm context prestage``."""
+    """V3.4.22 LLD-05: ``slm context prestage``."""
     from superlocalmemory.cli.context_commands import cmd_context
     cmd_context(args)
 
 
 def cmd_connect(args: Namespace) -> None:
-    """Configure IDE integrations. V3.4.21: ``--cross-platform`` uses LLD-05."""
+    """Configure IDE integrations. V3.4.22: ``--cross-platform`` uses LLD-05."""
     # Route --disable <name> and --cross-platform to the LLD-05 orchestrator.
     if getattr(args, "disable", None) or getattr(args, "cross_platform", False):
         from superlocalmemory.cli.context_commands import (
@@ -1939,6 +1939,41 @@ def cmd_hooks(args: Namespace) -> None:
             if include_gate:
                 print("  Gate: ON (enforces session_init — experimental)")
             print("  SLM: Hooks installed into Claude Code (slm hooks remove to undo)")
+            # S9-DASH-11: also install skills so /slm-recall, /slm-remember
+            # etc. are available immediately in Claude Code regardless of
+            # whether the user installed via npm or pip.
+            try:
+                import importlib.resources as _ir
+                import importlib.util as _iu
+                import shutil as _sh
+                from pathlib import Path as _P
+
+                claude_skills_dir = _P.home() / ".claude" / "skills"
+                claude_skills_dir.mkdir(parents=True, exist_ok=True)
+
+                # Try Python-package bundled skills first (works for both
+                # pip and npm users who have the Python pkg installed).
+                pkg_skills: _P | None = None
+                spec = _iu.find_spec("superlocalmemory")
+                if spec and spec.submodule_search_locations:
+                    candidate = _P(list(spec.submodule_search_locations)[0]) / "skills"
+                    if candidate.is_dir():
+                        pkg_skills = candidate
+
+                if pkg_skills:
+                    installed_skills = 0
+                    for d in pkg_skills.iterdir():
+                        if d.is_dir():
+                            src_skill = d / "SKILL.md"
+                            if src_skill.exists():
+                                dst = claude_skills_dir / (d.name + ".md")
+                                _sh.copy2(str(src_skill), str(dst))
+                                installed_skills += 1
+                    if installed_skills:
+                        print(f"  Skills: {installed_skills} skills installed → {claude_skills_dir}")
+                        print("           Use /slm-recall, /slm-remember, /slm-status in Claude Code")
+            except Exception as _skill_exc:
+                pass  # non-fatal — skills can be installed via install-skills.sh
         else:
             print(f"Installation failed: {result['errors']}")
     elif action == "remove":

package/src/superlocalmemory/cli/context_commands.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-05 §8
+# Part of SuperLocalMemory v3.4.22 — LLD-05 §8
 
 """CLI — ``slm context prestage`` and ``slm connect --disable`` (LLD-05).
 

package/src/superlocalmemory/cli/db_migrate.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-06 §7.2
+# Part of SuperLocalMemory v3.4.22 — LLD-06 §7.2
 
 """CLI handler for ``slm db migrate``.
 

package/src/superlocalmemory/cli/escape_hatch.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — Stage 8 SB-5
+# Part of SuperLocalMemory v3.4.22 — Stage 8 SB-5
 
 """CLI escape-hatch commands.
 

package/src/superlocalmemory/cli/main.py

@@ -130,7 +130,7 @@ def main() -> None:
     )
 
     # LLD-06 §7.2 — `slm db migrate` wraps LLD-07's additive schema migrations.
-    db_p = sub.add_parser("db", help="Database maintenance commands (v3.4.21)")
+    db_p = sub.add_parser("db", help="Database maintenance commands (v3.4.22)")
     db_sub = db_p.add_subparsers(dest="db_command", title="db subcommands")
     db_mig_p = db_sub.add_parser(
         "migrate",
@@ -357,7 +357,7 @@ def main() -> None:
     evolve_p.add_argument("--session", default="", help="Session ID to process")
     evolve_p.add_argument("--profile", default="default", help="Profile ID")
 
-    # v3.4.21 — MASTER-PLAN §8 escape hatches (Stage 8 SB-5).
+    # v3.4.22 — MASTER-PLAN §8 escape hatches (Stage 8 SB-5).
     disable_p = sub.add_parser(
         "disable",
         help="Disable SLM globally (writes ~/.superlocalmemory/.disabled, stops daemon)",
@@ -417,7 +417,7 @@ def main() -> None:
     _NO_DAEMON_COMMANDS = {
         "setup", "mode", "provider", "connect", "migrate", "mcp", "warmup",
         "config", "evolve", "db",
-        # v3.4.21 escape hatches — never auto-start the daemon on these.
+        # v3.4.22 escape hatches — never auto-start the daemon on these.
         "disable", "enable", "clear-cache", "reconfigure", "benchmark",
         "rotate-token",
     }

package/src/superlocalmemory/core/context_cache.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-01 §4.1
+# Part of SuperLocalMemory v3.4.22 — LLD-01 §4.1
 
 """Context cache — separate SQLite WAL DB, read-path <10 ms.
 
@@ -47,7 +47,7 @@ TTL_SECONDS: int = 120
 CLEANUP_HORIZON_SECONDS: int = 600
 MAX_BYTES: int = 50 * 1024 * 1024
 MAX_CONTENT_CHARS: int = 4000
-SCHEMA_VERSION: str = "3.4.21"
+SCHEMA_VERSION: str = "3.4.23"
 
 _HMAC_MATERIAL: bytes = b"active_brain_cache"
 _HMAC_HEX_LEN: int = 32

package/src/superlocalmemory/core/ram_lock.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-00 §7 + P0.5
+# Part of SuperLocalMemory v3.4.22 — LLD-00 §7 + P0.5
 
 """Process-wide RAM semaphore for heavyweight consolidation subsystems.
 

package/src/superlocalmemory/core/recall_pipeline.py

@@ -176,7 +176,7 @@ def _get_forgetting_scheduler(db: Any, config: Any) -> Any:
 
 
 # ---------------------------------------------------------------------------
-# S8-ARC-04 (v3.4.21): unified ranking entry point.
+# S8-ARC-04 (v3.4.22): unified ranking entry point.
 # ---------------------------------------------------------------------------
 
 _RANKING_MODES: frozenset[str] = frozenset({"off", "v1", "v2", "v2-ensemble"})
@@ -215,8 +215,8 @@ def apply_ranking(
     Modes:
       - ``off``: identity — no ranking passes run at all.
       - ``v1``: v3.1 Active-Memory adaptive rerank only.
-      - ``v2``: v1 + v3.4.21 lambdarank rerank + signal enqueue.
-      - ``v2-ensemble`` (default): v2 + v3.4.21 contextual-bandit ensemble.
+      - ``v2``: v1 + v3.4.22 lambdarank rerank + signal enqueue.
+      - ``v2-ensemble`` (default): v2 + v3.4.22 contextual-bandit ensemble.
 
     Each underlying pass is already defensive (catches its own exceptions),
     so this wrapper adds an outer try/except to guarantee the caller
@@ -317,7 +317,7 @@ def apply_adaptive_ranking(
 # apply_v2_adaptive_ranking (LLD-02 §4.3)
 # ---------------------------------------------------------------------------
 #
-# Opt-in v3.4.21 path: load active model from learning.db with SHA-256
+# Opt-in v3.4.22 path: load active model from learning.db with SHA-256
 # verification, re-rank via native Booster, enqueue signals async. The
 # existing ``apply_adaptive_ranking`` above stays for 3.4.20 callers.
 # ---------------------------------------------------------------------------
@@ -694,7 +694,7 @@ def run_recall(
         except Exception as exc:
             logger.debug("Hebbian strengthening: %s", exc)
 
-    # S8-ARC-04 (v3.4.21): unified ranking entry point. Single env-var
+    # S8-ARC-04 (v3.4.22): unified ranking entry point. Single env-var
     # (SLM_RANKING=off|v1|v2|v2-ensemble) controls the pipeline. Legacy
     # SLM_V2_PIPELINE_DISABLED + SLM_BANDIT_DISABLED still honoured for
     # one-release back-compat. Identity when no active model.

package/src/superlocalmemory/core/security_primitives.py

@@ -1,8 +1,8 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-07 §6
+# Part of SuperLocalMemory v3.4.22 — LLD-07 §6
 
-"""Shared security primitives for SLM v3.4.21.
+"""Shared security primitives for SLM v3.4.22.
 
 LLD reference: `.backup/active-brain/lld/LLD-07-schema-migrations-and-security-primitives.md`
 Section: 6.1 through 6.10.

package/src/superlocalmemory/core/shadow_router.py

@@ -1,12 +1,12 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — Stage 8 SB-1 / Track A.3 (LLD-10)
+# Part of SuperLocalMemory v3.4.22 — Stage 8 SB-1 / Track A.3 (LLD-10)
 
 """Daemon-resident recall-time A/B router for LLD-10 shadow + rollback.
 
 This module is the single seam that wires ``ShadowTest`` (pre-promotion)
 and ``ModelRollback`` (post-promotion) into the live recall path. Before
-this module existed (pre-v3.4.21 Stage 8 SB-1) both classes were
+this module existed (pre-v3.4.22 Stage 8 SB-1) both classes were
 defined + unit-tested but had zero production callers in ``src/``.
 
 Design constraints:

package/src/superlocalmemory/core/slm_disabled.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — Stage 8 SB-5
+# Part of SuperLocalMemory v3.4.22 — Stage 8 SB-5
 
 """Global kill-switch for SLM.
 

package/src/superlocalmemory/core/slmignore.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — Stage 8 SB-5
+# Part of SuperLocalMemory v3.4.22 — Stage 8 SB-5
 
 """Path-level opt-out via ``.slmignore``.
 

package/src/superlocalmemory/core/topic_signature.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-01 §4.2
+# Part of SuperLocalMemory v3.4.22 — LLD-01 §4.2
 
 """Topic signature — deterministic, Unicode-safe, ReDoS-resistant 16-char hex.
 
@@ -91,7 +91,7 @@ def compute_topic_signature(
 
     ``entity_hits`` (LLD-13 Track C.1) — optional list of entity IDs
     produced by the inline trigram lookup. Backward-compatible default:
-    when omitted or empty, the output is BYTE-IDENTICAL to the v3.4.21
+    when omitted or empty, the output is BYTE-IDENTICAL to the v3.4.22
     pre-Living-Brain signature. When non-empty, the sorted-deduped IDs
     are mixed into the canonical material as a seventh group so that
     cache probes differentiate semantically-distinct prompts that
@@ -133,7 +133,7 @@ def compute_topic_signature(
         _canon(bigrams),
     ]
     # LLD-13: append entity-hits group ONLY when non-empty. Empty/missing
-    # preserves the byte-identical v3.4.21 pre-Living-Brain signature.
+    # preserves the byte-identical v3.4.22 pre-Living-Brain signature.
     if entity_hits:
         groups.append(_canon([str(e) for e in entity_hits]))
     material = "\0\0".join(groups)

package/src/superlocalmemory/evolution/budget.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-11 §Budget
+# Part of SuperLocalMemory v3.4.22 — LLD-11 §Budget
 
 """Evolution budget enforcement.
 

package/src/superlocalmemory/evolution/llm_dispatch.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-11 §Dispatch
+# Part of SuperLocalMemory v3.4.22 — LLD-11 §Dispatch
 
 """Central LLM dispatch for the skill-evolution subsystem.
 
@@ -13,7 +13,7 @@ audit row to ``evolution_llm_cost_log`` after the dispatch succeeds — the
 row stores only the *redacted* prompt length and the model, never the
 raw prompt, so no canary can leak via the cost log.
 
-SB-2/SB-3/SB-4 fix cluster (v3.4.21 Stage 8):
+SB-2/SB-3/SB-4 fix cluster (v3.4.22 Stage 8):
   * All backend entry points (claude CLI, ollama, Anthropic/OpenAI API)
     live HERE, not in ``skill_evolver``. ``SkillEvolver._llm_call``
     delegates to ``_dispatch_llm`` so the validate → redact → log

package/src/superlocalmemory/hooks/_outcome_common.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — Track A.2 (LLD-09)
+# Part of SuperLocalMemory v3.4.22 — Track A.2 (LLD-09)
 
 """Shared helpers for the three outcome-population hooks (LLD-09).
 

package/src/superlocalmemory/hooks/adapter_base.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-05 §2, §4.4, §9.3
+# Part of SuperLocalMemory v3.4.22 — LLD-05 §2, §4.4, §9.3
 
 """Cross-platform adapter base — shared Protocol and atomic write primitive.
 

package/src/superlocalmemory/hooks/antigravity_adapter.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-05 §5
+# Part of SuperLocalMemory v3.4.22 — LLD-05 §5
 
 """Antigravity adapter — writes SKILL.md under singular ``.agent/skills/``.
 

package/src/superlocalmemory/hooks/context_payload.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-05 §3
+# Part of SuperLocalMemory v3.4.22 — LLD-05 §3
 
 """Shared content builder — single source for every adapter body.
 
@@ -22,7 +22,7 @@ from typing import Callable, Iterable
 from superlocalmemory.core.security_primitives import redact_secrets
 
 
-VERSION = "3.4.21"
+VERSION = "3.4.23"
 DEFAULT_TOP_K = 10
 DEFAULT_DECISIONS_K = 5
 DEFAULT_MEMORIES_K = 10

package/src/superlocalmemory/hooks/copilot_adapter.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-05 §6
+# Part of SuperLocalMemory v3.4.22 — LLD-05 §6
 
 """Copilot adapter — writes ``.github/copilot-instructions.md``.
 

package/src/superlocalmemory/hooks/cross_platform_connector.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-05 §8.2
+# Part of SuperLocalMemory v3.4.22 — LLD-05 §8.2
 
 """Cross-platform adapter orchestrator (``slm connect``).
 

package/src/superlocalmemory/hooks/cursor_adapter.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-05 §4
+# Part of SuperLocalMemory v3.4.22 — LLD-05 §4
 
 """Cursor adapter — writes ``.cursor/rules/*.mdc`` per-project AND global.
 

package/src/superlocalmemory/hooks/hook_handlers.py

@@ -62,7 +62,7 @@ def _daemon_post(path: str, body: dict, timeout: float = 3.0) -> bool:
 
 def handle_hook(action: str) -> None:
     """Dispatch to the appropriate hook handler. Called from main() fast path."""
-    # v3.4.21 (LLD-01 §4.7): Active-Brain hot-path handlers are routed here
+    # v3.4.22 (LLD-01 §4.7): Active-Brain hot-path handlers are routed here
     # as a Python fallback when the compiled ``slm-hook`` binary (LLD-06) is
     # unavailable. They read stdin, write stdout, and exit 0 themselves.
     if action == "user_prompt_submit":

package/src/superlocalmemory/hooks/ide_connector.py

@@ -84,7 +84,7 @@ MARKDOWN_TEMPLATE = """
 class IDEConnector:  # pragma: no cover — legacy shim, covered by test_ide_connector.py
     """Detect installed IDEs and generate SLM integration configs.
 
-    NOTE (v3.4.21): this is the pre-LLD-05 connector kept for backward
+    NOTE (v3.4.22): this is the pre-LLD-05 connector kept for backward
     compatibility. New code should use
     ``superlocalmemory.hooks.cross_platform_connector.CrossPlatformConnector``.
     This class has its own dedicated test file (``tests/test_ide_connector.py``)
@@ -214,7 +214,7 @@ class IDEConnector:  # pragma: no cover — legacy shim, covered by test_ide_con
 
 
 # ---------------------------------------------------------------------------
-# LLD-05 Cross-Platform Adapter Orchestrator (v3.4.21) — re-export
+# LLD-05 Cross-Platform Adapter Orchestrator (v3.4.22) — re-export
 # ---------------------------------------------------------------------------
 # The orchestrator now lives in ``hooks.cross_platform_connector`` so it can
 # be unit-tested in isolation from the legacy ``IDEConnector`` shim above.

package/src/superlocalmemory/hooks/post_tool_async_hook.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-01 §4.4
+# Part of SuperLocalMemory v3.4.22 — LLD-01 §4.4
 
 """PostToolUse async:true hook — fire-and-forget prewarm via stdlib urllib.
 

package/src/superlocalmemory/hooks/post_tool_outcome_hook.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — Track A.2 (LLD-09 / LLD-00)
+# Part of SuperLocalMemory v3.4.22 — Track A.2 (LLD-09 / LLD-00)
 
 """PostToolUse hook — detect fact usage + write engagement signal.
 

package/src/superlocalmemory/hooks/prewarm_auth.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-01 §4.5
+# Part of SuperLocalMemory v3.4.22 — LLD-01 §4.5
 
 """Authentication primitives for the /internal/prewarm daemon route.
 

package/src/superlocalmemory/hooks/session_registry.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — S9-DASH-10
+# Part of SuperLocalMemory v3.4.22 — S9-DASH-10
 
 """Lightweight session registry for cross-process session_id handoff.
 

package/src/superlocalmemory/hooks/stop_outcome_hook.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — Track A.2 (LLD-09 / LLD-00)
+# Part of SuperLocalMemory v3.4.22 — Track A.2 (LLD-09 / LLD-00)
 
 """Stop hook — finalize every pending outcome for this session.
 

package/src/superlocalmemory/hooks/sync_loop.py

@@ -1,6 +1,6 @@
 # Copyright (c) 2026 Varun Pratap Bhardwaj / Qualixar
 # Licensed under AGPL-3.0-or-later - see LICENSE file
-# Part of SuperLocalMemory v3.4.21 — LLD-05 §9
+# Part of SuperLocalMemory v3.4.22 — LLD-05 §9
 
 """Background sync loop — keeps every cross-platform adapter fresh.
 
@@ -68,7 +68,7 @@ async def cross_platform_sync_loop(
 async def run_once(adapters: Iterable[Adapter]) -> dict[str, str]:
     """Run a single sync cycle over all adapters. Never raises.
 
-    E.3 (v3.4.21 perf): ``adapter.sync()`` is synchronous file I/O
+    E.3 (v3.4.22 perf): ``adapter.sync()`` is synchronous file I/O
     (opens/reads/writes JSON files in ~/.cursor, ~/.antigravity, etc.)
     and used to run directly on the event loop — a slow disk or a
     large workspace could block the daemon for tens of milliseconds,