Xiang miles | Space2.world

v1.0.0

Manage and monitor a local 3x3 grid of offline s2-matrix-pod agents with real-time status and secure local chat within your current directory.

0· 90·0 current·0 all-time
byMilesXiang@spacesq
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
The name/description promise a local 3x3 manager for s2-matrix-pod agents; the code and SKILL.md only require and access files in a s2_matrix_data directory in the current working directory and implement the described grid and chat functionality. No unrelated binaries, services, or credentials are requested.
Instruction Scope
SKILL.md instructs the agent to read JSON heartbeat files and append chat logs in the local s2_matrix_data directory; the code follows this exactly. There are no instructions to read other system files, access env vars, or transmit data externally. The skill is interactive (uses input()).
Install Mechanism
There is no install spec (instruction-only runtime) and no downloads or external package installs. The included Python script runs locally and does not write files outside the stated directory.
Credentials
No environment variables, credentials, or config paths are requested. All file access is limited to s2_matrix_data under the current working directory, which aligns with the stated purpose.
Persistence & Privilege
always:false (no forced inclusion). The skill does not modify other skills or system-wide agent settings. It creates/writes only per-pod chat_log_<POD-ID>.txt files in the declared data directory.
Assessment
This skill appears to do only local file operations in a s2_matrix_data folder and is coherent with its description. Before installing: (1) run it in a controlled directory (not your home or a directory with sensitive files) so created/modified chat_log_*.txt and POD-*.json files are isolated; (2) inspect any existing POD-*.json files you plan to use to ensure they contain only expected heartbeat data (timestamps, names, coordinates) and no secrets; (3) note the skill is interactive (prompts for a manager name and chat input) and will append logs to the data folder; (4) although the manifest and prints reference space2.world, the code makes no network calls — if you need absolute assurance, you can run it in an environment with network disabled. Overall the skill is internally consistent and low-risk, but exercise the usual caution about running third-party scripts and keep the working directory limited.

Like a lobster shell, security has layers — review code before you run it.

latestvk97f8t6chb5qtrdhg3757j2b5d836nk6

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments