superlocalmemory 4.0.3 → 4.0.4

package/CHANGELOG.md

@@ -5,6 +5,28 @@ 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).
 
+## [4.0.4] - 2026-08-15 — Optional Bounded Loops evidence bridge
+
+### Added
+- An optional MCP 2 bridge discovers the separately installed
+  `bounded-loops-mcp` producer at call time and negotiates the public
+  `bounded-loops.dev/slm-bridge/v1` capability rather than pinning a producer
+  package version.
+- `observe_bounded_loop_evidence(workspace)` explicitly imports sanitized
+  terminal graph evidence using the producer's `run_ref` address. It is
+  available in the default, `code`, `full`, and `power` MCP surfaces.
+- M041 stores typed, profile-scoped external evidence in `learning.db`; Living
+  Brain displays honest Bounded Loop observation totals and documentation
+  explains the independent-product contract.
+
+### Safety
+- The bridge accepts no caller-supplied executable or shell arguments, has a
+  five-second producer timeout, validates the exact v1 evidence shape, and
+  preserves `eligible_for_learning: false`.
+- External receipts never enter recall, ranking, routing, reward, or automatic
+  learning. Profile erasure removes them and a durable tombstone blocks stale
+  processes from recreating them.
+
 ## [4.0.3] - 2026-08-15 — Consented host integration upgrades
 
 ### Added

package/README.md

@@ -5,7 +5,7 @@
   </picture>
 </p>
 
-<h1 align="center">SuperLocalMemory V4.0.3</h1>
+<h1 align="center">SuperLocalMemory V4.0.4</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>v4.0.3</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>v4.0.4</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/v4.0.3-Current_Release-2ea44f?style=for-the-badge&logo=checkmarx&logoColor=white" alt="v4.0.3 — Current Release"/></a>
+  <a href="CHANGELOG.md"><img src="https://img.shields.io/badge/v4.0.4-Current_Release-2ea44f?style=for-the-badge&logo=checkmarx&logoColor=white" alt="v4.0.4 — 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>
@@ -62,7 +62,7 @@ SuperLocalMemory V4 combines conventional dense and lexical retrieval with graph
 
 **Memory with a sense of time.** SLM does not only store *what* an agent learned — it records *when*. Every fact carries ingestion timing and provenance; recall runs a dedicated temporal candidate channel alongside semantic, lexical, and associative retrieval; scenes and entity timelines reconstruct sequence; and the lifecycle lets neglected memory decay and self-archive instead of growing without bound. Time is a first-class ranking and lifecycle signal rather than a timestamp column an agent never reads — which is what lets a long-lived agent reason about how its context changed, not only what it currently holds.
 
-**What V4.0.3 ships.** Existing users can now preview and explicitly refresh SLM-owned host integrations with `slm upgrade-hosts`, while package installation and MCP startup no longer silently modify host hook configuration. V4 remains a *governed* local-first control plane — every canonical write is admitted, policy-authorized, tracked as durable per-store obligations (lexical, temporal, vector), and sealed by a completion manifest, so it is either fully applied or explicitly marked degraded, never silently half-done. Flagship surfaces in this release:
+**What V4.0.4 ships.** Existing users can now preview and explicitly refresh SLM-owned host integrations with `slm upgrade-hosts`, while package installation and MCP startup no longer silently modify host hook configuration. V4.0.4 also adds an optional, capability-negotiated Bounded Loops observation bridge: compatible terminal graph receipts enter a separate, profile-scoped `learning.db` table and appear in Living Brain, without affecting recall, ranking, routing, or automatic learning. V4 remains a *governed* local-first control plane — every canonical write is admitted, policy-authorized, tracked as durable per-store obligations (lexical, temporal, vector), and sealed by a completion manifest, so it is either fully applied or explicitly marked degraded, never silently half-done. Flagship surfaces in this release:
 
 - **[SLM-Mesh](#slm-mesh-cross-session--cross-machine-coordination)** — authenticated cross-session and cross-machine peer coordination (messages, locks, shared state, inbox/outbox, optional discovery). Coordination only — not automatic replicated memory.
 - **Multi-scope memory & profiles** — workspaces (profiles) plus `personal` / `shared` / `global` scopes; cross-profile recall is default-deny.
@@ -71,7 +71,7 @@ SuperLocalMemory V4 combines conventional dense and lexical retrieval with graph
 - **Modes A / B / C** — local-only (A), on-device LLM enrichment (B), provider-assisted (C). An operating mode records technical locality facts; it does **not** determine EU AI Act legal compliance (that is deployment-context assessment — see [Privacy controls](#privacy-controls-and-operating-modes)).
 - **GDPR posture, retention & audit chain** — export, fail-closed cross-store erasure, retention policies, and a hash-chained audit trail. Engineering controls for compliance programs, not a legal certification.
 - **7-layer retrieval/recall stack & code graph** — multi-channel candidates (semantic, BM25, temporal, Hopfield, spreading activation) plus optional code-graph tools for blast radius and review context.
-- **MCP profiles** — `code` exposes **28** tools for installed coding agents; `full` **46**; `power` **58**; `whole` **91** (all registered). Also `core` (14), `mesh` (8), and the unrestricted default surface (46 with mesh enabled).
+- **MCP profiles** — `code` exposes **29** tools for installed coding agents; `full` **47**; `power` **59**; `whole` **92** (all registered). Also `core` (14), `mesh` (8), and the unrestricted default surface (47 with mesh enabled).
 - **Governed write path & verifiable transactions** — admission + policy control, a per-owner obligation ledger, and a hash-sealed completion manifest with a reconciler that redrives unmet obligations.
 - **Self-healing lifecycle & admin remediation** — stale locks cleared on restart; list/resolve stuck operations from CLI, MCP, or the dashboard.
 
@@ -436,11 +436,11 @@ Control tool surface via `SLM_MCP_PROFILE`:
 | Profile | Tools | Use case |
 |:--------|:-----:|:---------|
 | `core` | 14 | Memory, session, and optimize core |
-| `code` | 28 | Core + portable Brain evidence + code-graph tools + profile switching + bounded loops |
+| `code` | 29 | Core + portable Brain evidence + code-graph tools + profile switching + bounded loops |
 | `mesh` | 8 | SLM-Mesh only — multi-session / multi-machine coordination |
-| `full` | 46 | Memory + portable Brain evidence + optimize + evolution + mesh + bounded loops |
-| `power` | 58 | Full + administration, lifecycle, and diagnostics |
-| `whole` | 91 | Every registered MCP tool |
+| `full` | 47 | Memory + portable Brain evidence + optimize + evolution + mesh + bounded loops |
+| `power` | 59 | Full + administration, lifecycle, and diagnostics |
+| `whole` | 92 | Every registered MCP tool |
 
 **Precedence:** `ALL` > `TOOLS` > `PROFILE` > `default`
 
@@ -451,7 +451,7 @@ slm mcp
 
 For a predictable small surface, set `core` explicitly. Leaving the variable
 unset retains the compatibility default, whose mesh tools follow the local
-mesh setting. Count-suffixed aliases remain for backward compatibility and emit a migration warning: `core14`, `code20`, `code21`, `code24`, `code28`, `mesh8`, `full38`, `full39`, `full42`, `full46`, `power50`, `power51`, `power54`, `power58`, `whole81`, `whole84`, `whole91`. Unknown names stop startup instead of silently selecting another tool set.
+mesh setting. Count-suffixed aliases remain for backward compatibility and emit a migration warning: `core14`, `code20`, `code21`, `code24`, `code28`, `code29`, `mesh8`, `full38`, `full39`, `full42`, `full46`, `full47`, `power50`, `power51`, `power54`, `power58`, `power59`, `whole81`, `whole84`, `whole91`, `whole92`. Unknown names stop startup instead of silently selecting another tool set.
 
 Per-IDE configs available for Claude Code, Cursor, Windsurf, VS Code Copilot, Continue, Gemini CLI, JetBrains, Zed, and more (15 configs in `ide/configs/`). See [docs/ide-setup.md](docs/ide-setup.md).
 
@@ -684,6 +684,7 @@ interface.
 | Published benchmark evidence | [docs/benchmarks.md](docs/benchmarks.md) |
 | CLI reference | [docs/cli-reference.md](docs/cli-reference.md) |
 | MCP tools reference | [docs/mcp-tools.md](docs/mcp-tools.md) |
+| Optional Bounded Loops bridge | [docs/bounded-loops-bridge.md](docs/bounded-loops-bridge.md) |
 | Getting started | [docs/getting-started.md](docs/getting-started.md) |
 | IDE setup (15 configs) | [docs/ide-setup.md](docs/ide-setup.md) |
 | Teams, users, and RBAC | [docs/rbac-teams.md](docs/rbac-teams.md) |

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "superlocalmemory",
-  "version": "4.0.3",
+  "version": "4.0.4",
   "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": "4.0.3"
+  "version": "4.0.4"
 }

package/plugin/CLAUDE.md

@@ -1,4 +1,4 @@
-<!-- BEGIN SuperLocalMemory v4.0.3 -->
+<!-- BEGIN SuperLocalMemory v4.0.4 -->
 
 ## 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 v4.0.3 -->
+<!-- END SuperLocalMemory v4.0.4 -->
 
-SuperLocalMemory v4.0.3 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.4 · 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 v4.0.3 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later

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

@@ -68,4 +68,4 @@ assessment. The gate is the authority.
 
 ---
 
-SuperLocalMemory v4.0.3 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.4 · 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 v4.0.3 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.4 · 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 v4.0.3 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later

package/plugin/requirements.txt

@@ -1 +1 @@
-superlocalmemory==4.0.3
+superlocalmemory==4.0.4

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 v4.0.3 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.4 · 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 v4.0.3 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.4 · 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 v4.0.3 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later*

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

@@ -311,4 +311,4 @@ profile. See `slm-profile` for the full profile switching workflow.
 
 ---
 
-SuperLocalMemory v4.0.3 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.4 · 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 v4.0.3 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later

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

@@ -279,4 +279,4 @@ mesh availability.
 
 ---
 
-*SuperLocalMemory v4.0.3 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later*

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

@@ -31,9 +31,9 @@ project, a client engagement, a production vs staging environment.
 | Profile | Tools | When to use |
 |---------|-------|-------------|
 | `core` | 14 tools — remember, recall, search, session, optimize | Minimal footprint, no code tools |
-| `code` | 28 tools — core + portable Brain evidence + code graph + profile switching + bounded loops | Default for IDE/coding agents |
-| `full` | 46 tools — code + all memory ops + mesh + bounded loops | Multi-session, team workflows |
-| `power` | 58 tools — full + governance + behavioral tools | Enterprise, admin, audit use cases |
+| `code` | 29 tools — core + portable Brain evidence + code graph + profile switching + bounded loops | Default for IDE/coding agents |
+| `full` | 47 tools — code + all memory ops + mesh + bounded loops | Multi-session, team workflows |
+| `power` | 59 tools — full + governance + behavioral tools | Enterprise, admin, audit use cases |
 | `mesh` | 8 tools — mesh coordination only | Lightweight cross-session signalling |
 
 The profile is set at MCP server startup via `SLM_MCP_PROFILE` in the MCP config.
@@ -145,4 +145,4 @@ Name them differently in your MCP config (e.g. `superlocalmemory-personal` and
 
 ---
 
-*SuperLocalMemory v4.0.3 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.4 · 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 v4.0.3 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.4 · 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 v4.0.3 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.4 · 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 v4.0.3 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.4 · 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 v4.0.3 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later*

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

@@ -163,4 +163,4 @@ multi-profile setup. To switch the active profile, see `slm-profile`.
 
 ---
 
-SuperLocalMemory v4.0.3 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later

package/plugin-src/rules/AGENTS.md

@@ -82,10 +82,10 @@ When the SLM MCP server is unavailable, use these CLI equivalents:
 
 ## Tool reference (core profile — 14 tools)
 
-> The MCP config ships `SLM_MCP_PROFILE=code` (28 tools): the 14 core memory tools below
+> The MCP config ships `SLM_MCP_PROFILE=code` (29 tools): the 14 core memory tools below
 > **plus** 6 code-graph tools (`build_code_graph`, `get_blast_radius`, `query_graph`,
 > `semantic_search_code`, `get_review_context`, `detect_changes`) and `switch_profile`.
-> Use `full` (46 tools) to add mesh coordination. Use `power` (58 tools) for governance
+> Use `full` (47 tools) to add mesh coordination. Use `power` (59 tools) for governance
 > and audit tools. See slm-profile for profile switching.
 
 | Tool               | Signature (key params)                                                                       | Notes                                  |
@@ -128,4 +128,4 @@ 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 v4.0.3 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later

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

@@ -145,4 +145,4 @@ These subcommands control daemon-level cache settings. They do not read or write
 
 ---
 
-SuperLocalMemory v4.0.3 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later

package/plugin-src/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 v4.0.3 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later

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

@@ -245,4 +245,4 @@ Before running any destructive operation (`forget`, `compact_memories`):
 
 ---
 
-*SuperLocalMemory v4.0.3 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later*

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

@@ -311,4 +311,4 @@ profile. See `slm-profile` for the full profile switching workflow.
 
 ---
 
-SuperLocalMemory v4.0.3 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later

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

@@ -96,4 +96,4 @@ paused, name the approval needed; when errored, quote the short detail.
 
 ---
 
-SuperLocalMemory v4.0.3 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later

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

@@ -279,4 +279,4 @@ mesh availability.
 
 ---
 
-*SuperLocalMemory v4.0.3 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later*

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

@@ -31,9 +31,9 @@ project, a client engagement, a production vs staging environment.
 | Profile | Tools | When to use |
 |---------|-------|-------------|
 | `core` | 14 tools — remember, recall, search, session, optimize | Minimal footprint, no code tools |
-| `code` | 28 tools — core + portable Brain evidence + code graph + profile switching + bounded loops | Default for IDE/coding agents |
-| `full` | 46 tools — code + all memory ops + mesh + bounded loops | Multi-session, team workflows |
-| `power` | 58 tools — full + governance + behavioral tools | Enterprise, admin, audit use cases |
+| `code` | 29 tools — core + portable Brain evidence + code graph + profile switching + bounded loops | Default for IDE/coding agents |
+| `full` | 47 tools — code + all memory ops + mesh + bounded loops | Multi-session, team workflows |
+| `power` | 59 tools — full + governance + behavioral tools | Enterprise, admin, audit use cases |
 | `mesh` | 8 tools — mesh coordination only | Lightweight cross-session signalling |
 
 The profile is set at MCP server startup via `SLM_MCP_PROFILE` in the MCP config.
@@ -145,4 +145,4 @@ Name them differently in your MCP config (e.g. `superlocalmemory-personal` and
 
 ---
 
-*SuperLocalMemory v4.0.3 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later*

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

@@ -236,4 +236,4 @@ before recalling, then switch back. See `slm-profile` for workspace switching.
 
 ---
 
-*SuperLocalMemory v4.0.3 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later*

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

@@ -238,4 +238,4 @@ different workspace, use `switch_profile` first. See `slm-profile`.
 
 ---
 
-*SuperLocalMemory v4.0.3 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later*

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

@@ -173,4 +173,4 @@ to review the impact. See `slm-remember` for the full deletion discipline.
 
 ---
 
-*SuperLocalMemory v4.0.3 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later*

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

@@ -227,4 +227,4 @@ explicitly and call `recall` with `include_global`/`include_shared` after
 
 ---
 
-*SuperLocalMemory v4.0.3 · Qualixar · AGPL-3.0-or-later*
+*SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later*

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

@@ -163,4 +163,4 @@ multi-profile setup. To switch the active profile, see `slm-profile`.
 
 ---
 
-SuperLocalMemory v4.0.3 · Qualixar · AGPL-3.0-or-later
+SuperLocalMemory v4.0.4 · Qualixar · AGPL-3.0-or-later

package/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "superlocalmemory"
-version = "4.0.3"
+version = "4.0.4"
 description = "Local-first agent memory with auditable hybrid retrieval"
 readme = "README.md"
 license = "AGPL-3.0-or-later"

package/src/superlocalmemory/__init__.py

@@ -32,7 +32,7 @@ if "OMP_NUM_THREADS" not in os.environ:
     os.environ["OMP_NUM_THREADS"] = "2"
 # ---------------------------------------------------------------------------
 
-__version__ = "4.0.3"
+__version__ = "4.0.4"
 
 _REQUIRED_VERSIONS = {
     "sentence_transformers": "5.3.0",

package/src/superlocalmemory/integrations/bounded_loops_mcp.py

@@ -0,0 +1,184 @@
+"""Versioned, read-only contract boundary for Bounded Loops MCP evidence."""
+
+from __future__ import annotations
+
+import asyncio
+import json
+import os
+import shutil
+import stat
+from collections.abc import Awaitable, Callable
+from copy import deepcopy
+from datetime import timedelta
+from pathlib import Path
+from typing import Any
+
+CONTRACT_ID = "bounded-loops.dev/slm-bridge/v1"
+_OBSERVATION_TIMEOUT_SECONDS = 5.0
+_MAX_MCP_TEXT_BYTES = 2 * 1024 * 1024
+_ADVERTISEMENT = {
+    "id": CONTRACT_ID,
+    "tool": "bl_graph_evidence",
+    "operation": "observe_terminal_run",
+}
+
+
+class BridgeUnavailable(ValueError):
+    """The installed producer does not advertise a compatible bridge contract."""
+
+
+def supports_bridge(capabilities: dict[str, Any]) -> bool:
+    """Negotiate on the declared public contract, never producer semver."""
+    advertised = capabilities.get("evidence_contracts")
+    return isinstance(advertised, list) and any(
+        isinstance(item, dict)
+        and all(item.get(key) == value for key, value in _ADVERTISEMENT.items())
+        for item in advertised
+    )
+
+
+def bridge_payload(evidence: dict[str, Any], *, profile_id: str) -> dict[str, Any]:
+    """Attach active-profile identity after refusing incompatible evidence."""
+    if evidence.get("contract") != CONTRACT_ID:
+        raise BridgeUnavailable("unsupported bounded-loops evidence contract")
+    if evidence.get("eligible_for_learning") is not False:
+        raise BridgeUnavailable("bounded-loops evidence is not observation-only")
+    # The producer has organisation/project metadata for its own control plane.
+    # SLM stores only the v1 observation receipt needed by its profile-scoped
+    # learning database; retaining arbitrary producer extensions would turn a
+    # versioned contract into an unbounded schema sink.
+    fields = (
+        "contract",
+        "workspace_id",
+        "run_ref",
+        "run_id",
+        "outcome",
+        "run_state",
+        "demonstration",
+        "eligible_for_learning",
+        "terminal_at",
+        "graph_digest",
+        "plan_digest",
+        "policy_digest",
+        "receipt",
+        "nodes",
+    )
+    if any(field not in evidence for field in fields):
+        raise BridgeUnavailable("bounded-loops evidence is missing required v1 fields")
+    return {field: deepcopy(evidence[field]) for field in fields} | {"profile_id": profile_id}
+
+
+async def observe_terminal_runs(
+    call_tool: Callable[[str, dict[str, Any]], Awaitable[dict[str, Any]]], *, profile_id: str
+) -> list[dict[str, Any]]:
+    """Collect only producer-advertised terminal evidence over an injected MCP transport."""
+    discovery = await call_tool("bl_capabilities", {})
+    if discovery.get("status") != "ok" or not supports_bridge(discovery.get("capabilities", {})):
+        raise BridgeUnavailable("bounded-loops does not advertise slm-bridge/v1")
+    listing = await call_tool("bl_graph_terminal_runs", {"limit": 100})
+    if listing.get("status") != "ok" or listing.get("contract") != CONTRACT_ID:
+        raise BridgeUnavailable("bounded-loops terminal listing is unavailable")
+    runs = listing.get("runs")
+    if not isinstance(runs, list):
+        raise BridgeUnavailable("bounded-loops terminal listing is malformed")
+    # The producer's limit is advisory.  Keep this explicit operation bounded
+    # even against a compatible but faulty/malicious producer.
+    runs = runs[:100]
+    observed: list[dict[str, Any]] = []
+    for run in runs:
+        if not isinstance(run, dict) or not isinstance(run.get("run_ref"), str):
+            raise BridgeUnavailable("bounded-loops terminal listing is malformed")
+        response = await call_tool("bl_graph_evidence", {"run_ref": run["run_ref"]})
+        if response.get("status") == "unavailable":
+            continue
+        if response.get("status") != "ok" or not isinstance(response.get("evidence"), dict):
+            raise BridgeUnavailable("bounded-loops evidence response is malformed")
+        observed.append(bridge_payload(response["evidence"], profile_id=profile_id))
+    return observed
+
+
+async def observe_from_stdio(*, command: str, cwd: str, profile_id: str) -> list[dict[str, Any]]:
+    """Run one bounded, explicit MCP 2 observation; never call from recall or remember."""
+    executable, workspace = Path(command), Path(cwd)
+    if (
+        not executable.is_absolute()
+        or not executable.is_file()
+        or not workspace.is_absolute()
+        or not workspace.is_dir()
+        or workspace.is_symlink()
+    ):
+        raise BridgeUnavailable(
+            "bounded-loops bridge requires an approved executable and workspace"
+        )
+    try:
+        executable = executable.resolve(strict=True)
+        workspace = workspace.resolve(strict=True)
+        mode = executable.stat().st_mode
+    except OSError as exc:
+        raise BridgeUnavailable("bounded-loops bridge path is unavailable") from exc
+    if not stat.S_ISREG(mode) or (
+        os.name != "nt" and mode & (stat.S_IWGRP | stat.S_IWOTH)
+    ):
+        raise BridgeUnavailable("bounded-loops executable is not a trusted regular file")
+    if executable.stat().st_uid not in {0, os.geteuid()}:
+        raise BridgeUnavailable("bounded-loops executable owner is not trusted")
+
+    from mcp import ClientSession, StdioServerParameters
+    from mcp.client.stdio import stdio_client
+
+    try:
+        parameters = StdioServerParameters(
+            command=str(executable), args=[], cwd=str(workspace)
+        )
+        async with stdio_client(parameters) as (read, write):
+            async with ClientSession(
+                read,
+                write,
+                read_timeout_seconds=timedelta(seconds=_OBSERVATION_TIMEOUT_SECONDS),
+            ) as session:
+                async def observe() -> list[dict[str, Any]]:
+                    await session.initialize()
+
+                    async def call(name: str, arguments: dict[str, Any]) -> dict[str, Any]:
+                        result = await session.call_tool(name, arguments)
+                        if result.isError:
+                            raise BridgeUnavailable(
+                                "bounded-loops rejected the observation request"
+                            )
+                        texts = [item.text for item in result.content if hasattr(item, "text")]
+                        if len(texts) != 1 or len(texts[0].encode("utf-8")) > _MAX_MCP_TEXT_BYTES:
+                            raise BridgeUnavailable("bounded-loops returned an invalid MCP payload")
+                        try:
+                            payload = json.loads(texts[0])
+                        except json.JSONDecodeError as exc:
+                            raise BridgeUnavailable("bounded-loops returned invalid JSON") from exc
+                        if not isinstance(payload, dict):
+                            raise BridgeUnavailable("bounded-loops returned an invalid MCP payload")
+                        return payload
+
+                    return await observe_terminal_runs(call, profile_id=profile_id)
+                return await asyncio.wait_for(observe(), timeout=_OBSERVATION_TIMEOUT_SECONDS)
+    except BridgeUnavailable:
+        raise
+    except Exception as exc:
+        raise BridgeUnavailable("bounded-loops observation timed out or could not start") from exc
+
+
+async def observe_installed(*, workspace: str, profile_id: str) -> list[dict[str, Any]]:
+    """Observe a user-installed producer without accepting an agent command.
+
+    Discovery deliberately resolves exactly the public ``bounded-loops-mcp``
+    executable.  It does not accept a command, shell fragment, or arguments
+    from an MCP caller; the only caller-supplied value is the existing project
+    workspace whose Bounded Loops state is to be read.
+    """
+    command = shutil.which("bounded-loops-mcp")
+    if command is None:
+        raise BridgeUnavailable("bounded-loops-mcp is not installed")
+    if not Path(command).is_absolute():
+        raise BridgeUnavailable("bounded-loops-mcp discovery returned an unsafe path")
+    if Path(command).resolve().name not in {"bounded-loops-mcp", "bounded-loops-mcp.exe"}:
+        raise BridgeUnavailable("bounded-loops-mcp discovery returned an unsafe executable")
+    return await observe_from_stdio(
+        command=str(Path(command).resolve()), cwd=workspace, profile_id=profile_id
+    )

package/src/superlocalmemory/learning/database.py

@@ -613,7 +613,8 @@ class LearningDatabase:
                     row[0]
                     for row in conn.execute(
                         "SELECT name FROM sqlite_master WHERE type='table' "
-                        "AND name IN ('agent_experiences', 'cognitive_turn_receipts')"
+                        "AND name IN ('agent_experiences', 'cognitive_turn_receipts', "
+                        "'external_evidence_receipts')"
                     )
                 }
                 if profile_id is None: