PabloTheThinker · PabloTheThinker · Aug 3, 2026
diff --git a/ABOUT.md b/ABOUT.md
@@ -1,4 +1,4 @@
-**Hermespace is a functional J-Space workbench for Hermes agents** — FOA desk, dual decode, and an append-only world projection. Not [J-Space](https://github.com/anomalyco/j-space) neural weights. Not a second runtime. A harness-level global workspace inside Hermes that runs **standalone** and is **powered by [HermesCube](https://github.com/PabloTheThinker/hermescube)** when present.
+**Hermespace is a functional J-Space workbench for Hermes agents** — FOA desk, dual decode, and an append-only world projection. Not neural-weight J-space. Not a second runtime. A harness-level global workspace inside Hermes that runs **standalone** and is **powered by [HermesCube](https://github.com/PabloTheThinker/hermescube)** when present.
 
 Alongside the desk: skills+MEMORY fabric, neural FOA, autonomy grid, Cube heart/center cable (soft-fail).
 
@@ -37,11 +37,11 @@ Hermespace solves this with two systems:
 
 | Project | Relation |
 |---------|----------|
-| [ActiveGraph](https://github.com/anomalyco/ActiveGraph) | Event log is the agent, graph is the world — influenced the archive-first design |
-| [J-Space](https://github.com/anomalyco/j-space) | Verbal workspace, ~25 concepts, broadcasting hub — inspired the `concepts` system and `_refresh_concepts()` |
+| Anthropic J-space research | Verbal workspace, reportability, selective mediation — inspired desk concepts and dual decode |
 | [Missing Knowledge Layer](https://arxiv.org/abs/2405.10697) | Knowledge = supersession, Memory = decay, Wisdom = evidence-gated — inspired the epoch progression from unfiltered archive → evidence-gated wisdom |
 | [Vygotsky](https://en.wikipedia.org/wiki/Inner_speech) | Inner speech, signs as instruments of thought — influenced the dual decode architecture (report to human vs context to model) |
 | [Hermes Agent](https://github.com/NousResearch/hermes-agent) | The agent that grows with you, by Nous Research |
+| [HermesCube](https://github.com/PabloTheThinker/hermescube) | Optional heart/center cable — powers dream/warehouse path when present |
 
 ---
 

diff --git a/FOR_HERMES.md b/FOR_HERMES.md
@@ -65,7 +65,7 @@ When the plugin directory is a **symlink into this checkout**, import auto-resol
 ## Autonomy grid (v0.14)
 
 Missions, lenses, dream, self-talk, skillbench (hot-swap / merge / mutate), title tree.  
-Ground-up for Hermespace — not a port of AgentDrive or Conductor.
+Native Hermespace grid — built for this surface.
 
 ```bash
 hs grid status

diff --git a/README.md b/README.md
@@ -27,7 +27,7 @@
 
 **Hermespace is an append-only persistent world for Hermes agents.** Every session, every belief, every landmark, every evolution is recorded in an archive that never prunes, never decays, never caps. The agent builds a deepening model of itself and its environment across sessions — and it outlives the user.
 
-Not [J-Space](https://github.com/anomalyco/j-space). Not a second agent runtime. A room inside Hermes that remembers everything.
+Not a neural-weight J-space. Not a second agent runtime. A room inside Hermes that remembers everything.
 
 ---
 

diff --git a/desktop_plugin/hermespace/README.md b/desktop_plugin/hermespace/README.md
@@ -8,7 +8,7 @@
 ## Surfaces (v0.17.1)
 | Surface | How |
 |---------|-----|
-| **Pane** `hermespace` | `area: panes` — dockable like ilo-ops |
+| **Pane** `hermespace` | `area: panes` — dockable Hermes Desktop pane |
 | **Page** `/hermespace` | `ROUTES_AREA` |
 | **Sidebar** Hermespace | `SIDEBAR_NAV_AREA` → openRouteTile |
 | **Chip** `hs` | statusBar → navigate page |

diff --git a/desktop_plugin/hermespace/plugin.js b/desktop_plugin/hermespace/plugin.js
@@ -10,7 +10,7 @@
  *   (symlink to another tree breaks some Desktop/remote read paths)
  *
  * Surfaces:
- *  1) panes tile — immediate, dockable (like ilo-ops; users find this)
+ *  1) panes tile — immediate, dockable (users find this)
  *  2) full page /hermespace + sidebar.nav (first-class)
  *  3) status chip + palette → navigate /hermespace
  */
@@ -722,7 +722,7 @@ export default {
   name: 'Hermespace',
   defaultEnabled: true,
   register: function (ctx) {
-    // 1) Dockable pane — same class as working ilo-ops (always visible if plugins load)
+    // 1) Dockable pane — always visible if plugins load
     ctx.register({
       id: 'pane',
       area: 'panes',

diff --git a/docs/04-pattern-matrix.md b/docs/04-pattern-matrix.md
@@ -8,15 +8,15 @@
 | `wm.episodic_buffer` | Baddeley 2000 episodic buffer | Bind multimodal chunks | `bind_episode → [bind] concept` | yes |
 | `wm.central_executive` | Baddeley central executive | Control: update/select/protect (inhibit, shift) | `executive_mode + do_not_say inhibit` | yes |
 | `gwt.competition` | Baars/Dehaene GWT ignition/competition | Salience competition for broadcast | `compete_for_focus by salience` | yes |
-| `gwt.broadcast` | GWT global broadcast + Conductor inject | Winning contents available to speech/control | `build_inject_block focus-first` | yes |
+| `gwt.broadcast` | GWT global broadcast + Hermespace inject | Winning contents available to speech/control | `build_inject_block focus-first` | yes |
 | `load.sweller` | Sweller cognitive load theory | Intrinsic / extraneous / germane | `classify_message_load` | yes |
 | `load.protect` | Monotropism + high load / EF tax (ops layer) | Under high load shrink options and menus | `executive=protect + inject compression` | yes |
 | `desk.verbalizable` | Anthropic J-space reportability | Contents poised for report/speech | `required say field; cli say` | yes |
 | `desk.modulable` | J-space modulation + attention set | Task shifts workspace contents | `enter CLEAR + recompute_cognition(user_message)` | yes |
 | `desk.pre_output` | J-space silent deliberation | Workspace before user-visible speech | `save_desk + pre_llm inject` | yes |
 | `gate.selectivity` | J-space selective mediation | Not all cognition uses workspace | `gate.should_inject` | yes |
-| `episodic.ring` | Conductor EpisodicStore | Capped event log | `EpisodicLog max 500` | yes |
-| `semantic.consolidate` | Conductor consolidate + semantic memory | Episodes → durable notes | `semantic.consolidate()` | yes |
+| `episodic.ring` | Episodic ring buffer (file-backed) | Capped event log | `EpisodicLog max 500` | yes |
+| `semantic.consolidate` | Episodic → semantic consolidation | Episodes → durable notes | `semantic.consolidate()` | yes |
 | `meta.tribe_encode` | Meta TRIBE v2 multi-stream encoding | Stimulus streams → neural prediction (reversed: → desk slots) | `streams.encode_stimulus text/audio/visual/production` | yes |
 | `meta.b2q_decode` | Meta Brain2Qwerty brain→text decode | Internal state → linguistic report | `streams.decode_to_report → Say draft` | yes |
 | `meta.production_hierarchy` | Meta thought-to-action language production | Intention → semantics → report | `production_stages in desk.meta` | yes |

diff --git a/docs/06-component-research.md b/docs/06-component-research.md
@@ -11,7 +11,7 @@ Each component mapped to a published pattern, then implemented.
 7. Selectivity — J-space-style gate  
 8. Multi-stream encode — Meta TRIBE-class reverse  
 9. Decode to report — Brain2Qwerty-class reverse  
-10. Episodic/semantic memory — Conductor-style stores  
+10. Episodic/semantic memory — file-backed stores  
 11. Eval harness — falsifiable metrics  
 
 **Gap:** neural Jacobian-lens readout (not implemented).
diff --git a/docs/14-workbench-pocket-dimension.md b/docs/14-workbench-pocket-dimension.md
@@ -9,7 +9,7 @@ It is the **internal room** a Hermes agent works in:
 - When an **order** arrives — run encode → desk/neural → decode user + model  
 - Between jobs — monotropic FOA; park the rest instead of thrashing  
 
-Same spirit as Conductor’s old “pocket dimension,” named and built as **Hermespace**.
+Built as **Hermespace** — the agent's durable internal room inside Hermes.
 
 ```
               ┌──────────────── Hermespace ────────────────┐
@@ -27,7 +27,7 @@ Same spirit as Conductor’s old “pocket dimension,” named and built as **He
 ```python
 from hermespace.workbench import Workbench
 
-wb = Workbench(agent_id="ilo", session_id="main")
+wb = Workbench(agent_id="my-agent", session_id="main")
 wb.enter()                          # into the pocket dimension
 wb.park_goal("Later: write docs")   # monotropic park
 wb.idle_tick()                      # while waiting
@@ -46,11 +46,11 @@ model_ctx = result["model_context"]
 CLI:
 
 ```bash
-hs workbench enter --agent-id ilo
+hs workbench enter --agent-id my-agent
 hs workbench park --goal "Later: docs"
 hs workbench idle
 hs workbench order -m "Fix auth" --goal "Fix auth" --say "On it." --force
-hs workbench status --agent-id ilo
+hs workbench status --agent-id my-agent
 ```
 
 ## Idle vs order

diff --git a/docs/18-autonomy-grid.md b/docs/18-autonomy-grid.md
@@ -1,7 +1,7 @@
 # Hermespace autonomy grid (v0.14)
 
-Ground-up grid inside the pocket dimension — **not** a port of AgentDrive or Conductor.
-Roles inspired by those systems; security and data models rebuilt for Hermespace.
+Ground-up grid inside the Hermespace pocket dimension.
+Security and data models are native to this surface.
 
 ## Thesis
 
@@ -77,18 +77,16 @@ print(g.context_block())  # also folded into inject when load not high
 | skills_guard idea | `gates.check_skill_promote` |
 | learning loop | merge/mutate + title tree |
 
-## AgentDrive / Conductor
+## Grid building blocks
 
-**Not vendored.** Principles remade:
-
-| Foreign idea | Hermespace rebuild |
-|--------------|-------------------|
+| Idea | Hermespace module |
+|------|-------------------|
 | Experience graph | missions + scars + dream log + study DB |
-| Growth merge / fused skills | `skill-merge` proposals |
-| Learned skills | `skill-mutate` proposals |
-| DNA/title | `title_tree` |
-| Overseer gates | `gates` budget + intent |
-| Pillars/scars | grid scars (typed), not Conductor plugin dual-load |
+| Skill fusion | `skill-merge` proposals |
+| Skill evolution | `skill-mutate` proposals |
+| Title / self-model | `title_tree` |
+| Autonomy gates | `gates` budget + intent |
+| Failure memory | grid scars (typed) |
 
 ## Meta-brain
 

diff --git a/docs/22-open-roadmap.md b/docs/22-open-roadmap.md
@@ -24,12 +24,10 @@ Snapshot after v0.17 desktop-detach + pulse API. Host-agnostic.
 |---|------|-----|--------|
 | 1 | **Event-driven pulse** (access_approved → immediate tick) | Faster than 1m wake | M |
 | 2 | **Propose-order path** (gated self-order when autonomy on) | True unattended loop | M |
-| 3 | **AgentDrive bridge** (optional experience write/read) | Long terrain without fork | L |
-| 4 | **Conductor adapter** (optional ethics/scar when present) | Unattended law | M |
-| 5 | **Fudoshin eval note** with/without grid under load | Research capital | M |
-| 6 | Desktop **start socket** helper from palette (spawn serve) | UX when offline | S |
-| 7 | Pulse job enable/disable toggles in Desktop UI | Operator control | S |
-| 8 | Multi-agent pulse boards in one page | Multi-agent hosts | M |
+| 3 | **Fudoshin eval note** with/without grid under load | Research capital | M |
+| 4 | Desktop **start socket** helper from palette (spawn serve) | UX when offline | S |
+| 5 | Pulse job enable/disable toggles in Desktop UI | Operator control | S |
+| 6 | Multi-agent pulse boards in one page | Multi-agent hosts | M |
 
 ## Open — polish
 
@@ -40,7 +38,7 @@ Snapshot after v0.17 desktop-detach + pulse API. Host-agnostic.
 
 ## Not doing (by design)
 
-- Porting AgentDrive/Conductor trees into Hermespace
+- Porting foreign product trees into Hermespace
 - Character kits as lenses
 - Unbounded autonomy / money / public post
 - Right-rail-only HS panel (removed — page is SoT)

diff --git a/docs/23-everyday-ops.md b/docs/23-everyday-ops.md
@@ -18,7 +18,6 @@ export HERMESPACE_ROOT=…/hermespace
 export HERMESPACE_HOME="${HERMESPACE_HOME:-$HOME/.hermespace}"
 export HERMES_HOME="${HERMES_HOME:-$HOME/.hermes}"
 export PYTHONPATH="$HERMESPACE_ROOT/src${PYTHONPATH:+:$PYTHONPATH}"
-# Optional ILO_HOME if desk lives under an operator home tree
 ```
 
 ## Daily boot (2 commands)

diff --git a/scripts/new_desk.sh b/scripts/new_desk.sh
@@ -1,6 +1,6 @@
 #!/usr/bin/env bash
 set -euo pipefail
-HOME_ROOT="${HERMESPACE_HOME:-${ILO_HOME:-$HOME/.hermespace}}"
+HOME_ROOT="${HERMESPACE_HOME:-$HOME/.hermespace}"
 DEST="$HOME_ROOT/memory/hermespace/ACTIVE.md"
 TEMPLATE="$(cd "$(dirname "$0")/.." && pwd)/runtime/ACTIVE.template.md"
 mkdir -p "$(dirname "$DEST")"

diff --git a/scripts/show_desk.sh b/scripts/show_desk.sh
@@ -1,6 +1,6 @@
 #!/usr/bin/env bash
 set -euo pipefail
-HOME_ROOT="${HERMESPACE_HOME:-${ILO_HOME:-$HOME/.hermespace}}"
+HOME_ROOT="${HERMESPACE_HOME:-$HOME/.hermespace}"
 DEST="$HOME_ROOT/memory/hermespace/ACTIVE.md"
 if [[ ! -f "$DEST" ]]; then
   echo "no active desk at $DEST"

diff --git a/skills/hermespace/SKILL.md b/skills/hermespace/SKILL.md
@@ -14,7 +14,7 @@ description: >-
 **What it is not:** A second LLM runtime · Claude J-space weights · a SaaS brand · a replacement for Hermes skills/MEMORY.
 
 **Package SoT:** `$HERMESPACE_ROOT` (git checkout) · https://github.com/PabloTheThinker/hermespace  
-**State:** `$HERMESPACE_HOME` default `~/.hermespace` (optional `ILO_HOME`)  
+**State:** `$HERMESPACE_HOME` default `~/.hermespace`  
 **Hermes:** `$HERMES_HOME` default `~/.hermes`  
 **Version:** align `src/hermespace/__init__.py` · `pyproject.toml` · `hermes_plugin/plugin.yaml`
 
@@ -373,7 +373,7 @@ Never end on only a verify table / `ADHOC_PASS`.
 - `hermespace-ops` · `hermespace-grid` · `hermespace-everyday` · `hermespace-runtime-ops`  
 - `pocket-dimension-security`  
 - `hermes-agent` · `hermes-desktop-surface` · `hermes-local-ops`  
-- Finish: `ilo-finish-report` / `material-finish-review` / `session-closeout` when those exist  
+- Finish: `session-closeout` / `material-finish-review` / `professional-messaging` when those exist  
 
 **Package skill path:** `$HERMESPACE_ROOT/skills/hermespace/SKILL.md`  
 **Profile install:** `$HERMES_HOME/skills/.../hermespace/SKILL.md` (via `install_hermes.sh`)

diff --git a/src/hermespace/episodic.py b/src/hermespace/episodic.py
@@ -1,4 +1,4 @@
-"""Episodic ring buffer for Hermespace events — strip inspired by Conductor EpisodicStore.
+"""Episodic ring buffer for Hermespace events.
 
 Capped newest-retained log so the workspace compounds without unbounded growth.
 """
@@ -31,7 +31,7 @@ class Episode:
 
 
 class EpisodicLog:
-    """File-backed episodic log (no Conductor SessionStore dependency)."""
+    """File-backed episodic log for Hermespace events."""
 
     def __init__(self, path: Path | None = None) -> None:
         self.path = path or (default_state_dir() / "episodes.jsonl")

diff --git a/src/hermespace/grid/gates.py b/src/hermespace/grid/gates.py
@@ -1,6 +1,6 @@
-"""Autonomy / safety gates — ground-up for Hermespace (not a Conductor port).
+"""Autonomy / safety gates for Hermespace.
 
-Principles borrowed as *ideas*, rebuilt for this surface:
+Principles:
 - Budget limits unattended action
 - Risk classes
 - No irreversible money/public without human

diff --git a/src/hermespace/grid/lenses.py b/src/hermespace/grid/lenses.py
@@ -55,7 +55,7 @@
     "partner": {
         "title": "Partner",
         "bias": "Monotropism, finish reports, high-load short say",
-        "fabric_boost": ["professional-messaging", "ilo-finish-report", "hermes-agent"],
+        "fabric_boost": ["professional-messaging", "session-closeout", "hermes-agent"],
         "inhibit": ["option menus under high load", "leave hanging with only verify table"],
         "report_style": "closeout",
     },

diff --git a/src/hermespace/grid/skillbench.py b/src/hermespace/grid/skillbench.py
@@ -1,10 +1,9 @@
 """Skillbench — hot-swap skill modules + merge/mutate evolution on the workbench.
 
-Ground-up (AgentDrive *roles* only: learned/fused skills, growth merge):
 - Modules live under HERMESPACE_HOME grid (not auto-installed to Hermes)
 - Merge two skill bodies → draft proposal
 - Mutate a skill with a delta note → draft proposal
-- Promote requires explicit call + static gate (Hermes skills_guard *idea*)
+- Promote requires explicit call + static gate
 """
 
 from __future__ import annotations
@@ -185,7 +184,7 @@ def merge_skills(
     new_name: str | None = None,
     note: str = "",
 ) -> Proposal:
-    """Fuse two module bodies into a draft (AgentDrive fused-* *role*)."""
+    """Fuse two module bodies into a draft proposal."""
     a = get_module_body(name_a, agent_id=agent_id)
     b = get_module_body(name_b, agent_id=agent_id)
     if not a or not b:

diff --git a/src/hermespace/paths.py b/src/hermespace/paths.py
@@ -11,13 +11,11 @@ def hermespace_home() -> Path:
 
     Order:
       HERMESPACE_HOME
-      ILO_HOME (compat)
       ~/.hermespace
     """
-    for key in ("HERMESPACE_HOME", "ILO_HOME"):
-        raw = os.environ.get(key, "").strip()
-        if raw:
-            return Path(raw).expanduser().resolve()
+    raw = os.environ.get("HERMESPACE_HOME", "").strip()
+    if raw:
+        return Path(raw).expanduser().resolve()
     return (Path.home() / ".hermespace").resolve()
 
 

diff --git a/src/hermespace/patterns.py b/src/hermespace/patterns.py
@@ -66,7 +66,7 @@ class Pattern:
     ),
     Pattern(
         "gwt.broadcast",
-        "GWT global broadcast + Conductor inject",
+        "GWT global broadcast + Hermespace inject",
         "Winning contents available to speech/control",
         "build_inject_block focus-first",
         True,
@@ -122,15 +122,15 @@ class Pattern:
     ),
     Pattern(
         "episodic.ring",
-        "Conductor EpisodicStore",
+        "Episodic ring buffer (file-backed)",
         "Capped event log",
         "EpisodicLog max 500",
         True,
         "episodic.py",
     ),
     Pattern(
         "semantic.consolidate",
-        "Conductor consolidate + semantic memory",
+        "Episodic → semantic consolidation",
         "Episodes → durable notes",
         "semantic.consolidate()",
         True,

diff --git a/src/hermespace/semantic.py b/src/hermespace/semantic.py
@@ -1,4 +1,4 @@
-"""Semantic notes — Conductor SemanticStore pattern, file-backed."""
+"""Semantic notes — durable statements consolidated from episodic events."""
 
 from __future__ import annotations
 
@@ -85,7 +85,7 @@ def add(self, statement: str, *, tags: list[str] | None = None, confidence: floa
 
 
 def consolidate(limit: int = 40) -> dict[str, Any]:
-    """Deterministic consolidation from episodic → semantic (Conductor pattern)."""
+    """Deterministic consolidation from episodic → semantic notes."""
     eps = EpisodicLog().recent(limit=limit)
     # recent() returns newest first; reverse for chrono
     entries = list(reversed(eps))

diff --git a/src/hermespace/workbench.py b/src/hermespace/workbench.py
@@ -7,8 +7,8 @@
 - **Idle tick** — maintain memory/neural while waiting for orders
 - **Order turn** — when a user/system order arrives, run through Hermespace
 
-This is the Conductor-era "pocket dimension" idea, realized as Hermespace
-state + API — not a separate product brand.
+Hermespace's pocket workbench: durable FOA desk + park stack + idle/order
+turns — state and API for Hermes agents, not a second product brand.
 """
 
 from __future__ import annotations
-Original file line number
+Diff line change
@@ Expand Up / @@ -27,7 +27,7 @@ @@
     **Hermespace is an append-only persistent world for Hermes agents.** Every session, every belief, every landmark, every evolution is recorded in an archive that never prunes, never decays, never caps. The agent builds a deepening model of itself and its environment across sessions — and it outlives the user.
-    Not [J-Space](https://github.com/anomalyco/j-space). Not a second agent runtime. A room inside Hermes that remembers everything.
+    Not a neural-weight J-space. Not a second agent runtime. A room inside Hermes that remembers everything.
     ---
@@ Expand Down @@