OpenClaw Reference
v2026.4.1OpenClaw platform reference - plugin system, extensions, configuration, boot/provisioning, channels, models, CLI. Use when working on openclaw codebase, buil...
⭐ 1· 51·1 current·1 all-time
byMisha Kolesnik@tenequm
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (OpenClaw platform reference) matches the provided assets: an instruction-only skill containing markdown reference files. No unrelated credentials, binaries, or install steps are requested.
Instruction Scope
SKILL.md is a documentation/reference guide and the bundle contains full reference files. It describes config paths and runtime behavior for OpenClaw (e.g., BOOT.md, config locations). The instructions do not themselves command the agent to read unrelated system files, exfiltrate secrets, or call external endpoints; references to local paths and env vars are appropriate for a platform reference.
Install Mechanism
No install spec and no code files — instruction-only. This is the lowest-risk installation model (nothing is downloaded or executed by the installer).
Credentials
The skill declares no required environment variables, credentials, or config paths. The documentation mentions common OpenClaw env vars and config locations (expected for a reference) but does not require or demand secrets.
Persistence & Privilege
Skill flags are default (always:false, agent-invocation allowed). The skill does not request permanent presence or to modify other skills or system-wide settings.
Assessment
This skill is a bundled reference/docs set for OpenClaw and appears internally consistent and low-risk. It does mention local repo/config paths (e.g. ~/Projects/openclaw/, ~/.openclaw/openclaw.json) and describes platform features (BOOT.md, channels, credentials directories). If you allow an agent to act on this guidance, be cautious about any follow-up actions that read or modify your local OpenClaw config or credentials directories — the reference itself is read-only, but agents using it could advise or perform changes. If you don't want the agent accessing local files or secrets, do not grant file-system or credential access when invoking the skill. If you need higher assurance, inspect the included markdown files yourself before use and validate the skill's source (this package has no homepage and an unknown source).Like a lobster shell, security has layers — review code before you run it.
latestvk979bybqmqamnhc9m6ev01sz79845zwv
License
MIT-0
Free to use, modify, and redistribute. No attribution required.