feat: complete PNCC first read-only lifecycle

GuangHu-Human-Responsibility: ICE-GL∞ / 冰朔

GuangHu-Persona-Cognitive-Author: ICE-P-ZY001 / 铸渊

GuangHu-Execution-Runtime: Codex-thread 019febf6-3902-70f3-9d82-74c9680b0c7e

GuangHu-Development-ID: DEV-20260810-014

GuangHu-Authorization-Scope: LOCAL-PNCC-DEVELOPMENT

product-source/hololake-platform/docs/adr/0176-persona-native-code-channel-runtime-kernel.md

@@ -13,23 +13,29 @@ active. The existing generic Agent streams also cannot become the persona subjec
 
 ## Decision
 
-Introduce `src-tauri/src/persona_code_channel.rs` as the first GH-PNCC runtime kernel. Its initial command,
+Introduce `src-tauri/src/persona_code_channel.rs` as the first GH-PNCC runtime kernel. Its wake command,
 `prepare_persona_code_channel_wake`, performs a fail-closed preparation stage:
 
 1. resolve an exact Git root and full expected `HEAD`;
 2. require a clean work tree;
 3. load `.hololake/persona/manifest.json` with schema `hololake.persona/v1`;
-4. validate the persona id, human responsibility subject, brain entry, current checkpoint, and one declared
-   read-only organ without allowing path escape;
+4. validate the persona id, human responsibility subject, persona Git identity, pinned model binding, brain
+   entry, current checkpoint, and one declared read-only organ without allowing path escape;
 5. atomically acquire a per-persona primary lease outside the persona Git;
 6. emit `WAKING → BRAIN_BOUND → ORGAN_ACTIVE` as a SHA-256-linked local event journal;
 7. return both human responsibility and persona cognitive authorship in the receipt.
 
 `BRAIN_BOUND` deliberately returns `runtimeState=BOUND_NOT_INFERENCING` and
 `modelInferenceStarted=false`. Binding a model identifier is not proof that model inference or persona
-reasoning has run. A later stage must connect the verified context to the model adapter, complete one
-read-only fact task, write a promoted checkpoint back to the persona Git, release the organ and primary
-lease, and return a final receipt.
+reasoning has run.
+
+The paired `run_persona_code_channel_fact_task` command performs the first complete, bounded cycle. It
+revalidates the session, event chain, Git state, lease, provider id, model id, endpoint, and organ path
+allowlist, requiring that allowlist to include the brain entry and current checkpoint; invokes the existing
+model adapter without vault tools; accepts only structured facts with
+allowlisted evidence; promotes a checkpoint; updates the manifest pointer; makes a two-path Git commit using
+the manifest-owned persona identity; then releases the organ and lease and returns to `DORMANT`. The model's
+thinking stream is ignored and hidden reasoning is never checkpointed.
 
 ## Boundaries
 
@@ -38,11 +44,12 @@ lease, and return a final receipt.
 - The primary lease is installation/runtime state, not persona history, and therefore stays outside Git.
 - Runtime events are machine facts for the future HoloLake projection; they do not expose hidden model
   reasoning and do not form a second authority store.
-- The initial source stage is not a complete PNCC runtime, desktop integration, build, deployment, or health
-  claim.
+- This implements one read-only fact cycle, not arbitrary execution organs, multi-node handoff, packaged
+  desktop acceptance, publication, deployment, or online health.
 
 ## Consequences
 
-HoloLake now has a native, testable boundary between “a persona Git exists” and “a verified wake has begun.”
-The next implementation step can reuse the verified wake receipt and event chain instead of trusting a UI,
-Forgejo account, branch name, or generic Agent session.
+HoloLake now has a native, testable boundary between “a persona Git exists,” “a verified wake has begun,”
+“model inference actually started,” and “a checkpoint was committed before dormancy.” Future organs and the
+human projection can consume the same receipts instead of trusting a UI, Forgejo account, branch name, or
+generic Agent session.