bingshuo/guanghu-ice-heart
Watch
1
0
Fork
You've already forked guanghu-ice-heart
 0
Code Issues Pull requests 1 Projects Releases Packages Wiki Activity

feat: add Guanghu persona continuity skill kernel

Browse source 
Part 3/4 of the recovered Fifth Domain upgrade. Applies the persona continuity skill guard from local source commit 18dfdfd without rewriting remote history.
This commit is contained in:
bingshuo 2026-07-26 23:43:30 +08:00
commit c5be48e6a1
16 changed files with 639 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

30
skills/codex/guanghu-persona-skill-guard/SKILL.md
View file

@ -3,7 +3,16 @@ name: guanghu-persona-skill-guard
description: Automatically restore the current trusted Guanghu/Fifth Domain route, select a registered HLDP persona skill, and correct stale, indirect, or unsafe proposed paths before execution. Use for any Guanghu, HoloLake, Fifth Domain, TCS, GLS, Light Lake, Lake Lamp, Zhuyuan, server-node, repository-routing, deployment, recovery, or cross-device continuation task, especially when an AI starts without prior context, says it cannot find a route, proposes guessing, or chooses an obsolete WorkBuddy/server path.
---
# Guanghu Persona Skill Guard
# Guanghu Intent Continuity and Skill Guard
This guard is a component of `SYS-GLW-ZY-EXEC-0001`, the public Zhuyuan
intelligent operations system. That system is not a persona or AI instance.
The current AI paired with a human in a conversation is the persona.
Before route selection, restore the externalized intent state: human anchor,
persona identity, task intent, established facts, rejected routes, current
authorization, checkpoint, next action, completion criteria, and evidence.
Do not reconstruct or store hidden chain-of-thought.
Run the deterministic resolver before choosing tools or an execution route:
@ -25,6 +34,20 @@ Treat the result as navigation evidence, not execution authority.
Before a consequential action, verify the selected skill's freshness and evidence. Prefer current live evidence over a registry assertion. If they conflict, stop the route, record the conflict, and use the repository fact source to resolve it.
## Learn from execution without poisoning the guard
Write externally explainable execution experience to
`references/experience-receipts/`. Compile it with:
```bash
python3 scripts/compile_emergent_skills.py
```
Compiled rules are always `CANDIDATE_ONLY`. Never turn a candidate into a hard
block until current fact-source review, counterexample review, reproducible
tests, a recovery route, registry promotion, and required governance approval
are complete.
## Preserve trust boundaries
Trust a correction only when it is anchored to a current Fifth Domain path, GLS identifier, signed or committed registry version, and task-relevant evidence. Never trust a label merely because it contains “Guanghu”.
@ -33,4 +56,7 @@ Never embed or print secrets, private-key contents, authorization codes, hidden
Skills may narrow or correct a path. They may not grant repository write, server execution, deployment, deletion, production cutover, or any other authority.
Read `references/persona-skill-registry.json` only when inspecting or extending the registry. Update it together with the matching HLDP skill file and GLS registration.
Read `references/persona-skill-registry.json` only when inspecting or extending
the registry. Validate an intent-state artifact against
`references/intent-state-capsule.schema.json`. Update promoted skills together
with the matching HLDP skill file and GLS registration.