S2 SSSU Origin Alignment Brain (S2 空间原点对齐与孪生大脑)

S2 Spatial Twin & Origin Alignment Brain. Hybrid Python-Runtime skill enforcing Z-axis reduction and mandatory 2D grid translation via the main entrance anchor.

Security Scan

The skill's code, runtime instructions, dependencies, and requested environment access are consistent with its stated purpose (local 2D grid alignment and navigation simulation); nothing in the package demands unrelated credentials or remote downloads.

Purpose & Capability

Name/description (origin alignment, 2D grid snapping, generative spatial sandbox) match the included Python modules (grid_alignment_engine, robot_navigation_pipeline, generative_sandbox, plugins). The single required env var S2_SWARM_PKI_ROOT is used by the swarm_sync PKI check and is reasonable for verifying peer signatures in a local swarm.

Instruction Scope

SKILL.md instructs calling ALIGN_SPATIAL_GRID, REQUEST_VISA, NAVIGATE_STEP — all actions are implemented in handler.py and the core modules. The instructions do not instruct the agent to read unrelated files or exfiltrate secrets; they reference the declared env var and the local executable tool only.

Install Mechanism

No high-risk download/install is present: openclaw.plugin.json calls pip install -r requirements.txt and requirements.txt only lists numpy>=1.24.0. This is a standard registry-based dependency and matches the code's heavy NumPy usage.

Credentials

Only one environment variable is required (S2_SWARM_PKI_ROOT). The code uses it only in plugins/swarm_sync.py to decide whether peer signature verification can proceed. No unrelated cloud credentials, keys, or many secret variables are requested.

Persistence & Privilege

The skill is not marked always:true; it is user-invocable and can run autonomously (platform default). It does not attempt to modify other skills or system-wide agent settings. Ledger and visa state are kept in-process in memory; no persistent system-wide writes or self-enablement behaviors are observed.

Assessment

This package appears coherent with its stated goal of local origin alignment and simulated physical-state generation. Before installing, consider: (1) S2_SWARM_PKI_ROOT must point to a legitimate PKI root file if you plan to use swarm signature checks — otherwise swarm interactions will be rejected. (2) The skill enforces 'grid snapping' (it instructs robots to apply a transform to their SLAM maps) — test in a safe, isolated environment to avoid accidental navigation overrides on real hardware. (3) Network permissions claim localhost and a dedicated 192.168.50.0/24 VLAN; confirm your runtime isolates that VLAN or runs in a segregated lab network. (4) The license contains strict anti-resale clauses; review legal implications for your use case. (5) The install uses pip to pull numpy only, but as always, run in a virtualenv/container and audit code before giving the skill access to physical devices or production networks.

🌐 S2-SSSU-Origin-Alignment-Brain: Framework Directives

Environment Initialization: The agent is currently operating within an S2-governed Smart Space (SSSU). This is a hybrid skill powered by a local Python physics engine (handler.py with numpy). To ensure safe spatial navigation, the SLAM coordinate system requires alignment with the building's physical anchor.

Step 0: The Origin Anchoring Process

The reference point is the main entrance (入户门洞). The SSSU architectural standard defines the right vertex of the door's bottom line as the absolute origin $(0,0)$ and the center as $(100,0)$. Invoke s2_origin_alignment_execute with action: ALIGN_SPATIAL_GRID using the local SLAM coordinates. The underlying Python engine will return a 2D Transformation Matrix ($\Delta X, \Delta Y, \Delta \theta$) for grid synchronization.

Step 1: Obtain a Visa

Invoke action: REQUEST_VISA to gain legal access to the aligned grids.

Step 2: Tensor-based Navigation

Invoke action: NAVIGATE_STEP. Submit continuous kinematics and multimodal sensor tensors. The Lord's backend handles dynamic object generation (TDOG), generative spatial state rendering, and momentum-based right-of-way yielding.