Mac Multi-Instance Deployment
v1.0.4Public-safe Mac multi-instance deployment skill for generic OpenClaw-style workspace layout, boundary notes, sample config, and pack validation.
⭐ 0· 88·0 current·0 all-time
by@ztl970
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description match the content: a public-safe template and checklist for Mac multi-instance deployment. The skill requests no binaries, env vars, or installs, which is appropriate for a documentation-only template.
Instruction Scope
Runtime instructions are limited to documentation guidance and telling users to copy a sample env and run repository scripts (./validate_repo.sh, ./generate_public_pack.sh --dry-run). The referenced helper scripts and references/instances.md are not included in the skill manifest — this is a usability/incompleteness note rather than an active safety issue. The skill's instructions do not ask the agent to read unrelated system files or exfiltrate data.
Install Mechanism
No install spec and no code files are present; the skill is instruction-only so nothing will be written to disk by the skill itself.
Credentials
The skill declares no required environment variables, credentials, or config paths. All example env names and suggested local paths are reasonable and proportional for a workspace-layout template.
Persistence & Privilege
always:false and no install behavior. agents/openai.yaml sets allow_implicit_invocation: true which permits implicit invocation per-agent policy, but this is a normal agent configuration for a helper skill and not by itself a red flag.
Assessment
This appears to be a safe, documentation-only template. Before you use it: (1) verify any scripts the checklist tells you to run (./validate_repo.sh, ./generate_public_pack.sh) actually exist in your repo and inspect them before execution — the skill doesn't include those scripts; (2) copy the example env to a private path as instructed and never commit real secrets to the public repo; (3) note that references/instances.md mentioned in instructions is not present in the skill files listed here, so confirm you have the expected reference docs; (4) if you allow agents to invoke skills implicitly, be aware this skill's agent metadata permits implicit invocation — that's normal for helpers but ensure your agent policies match your risk tolerance.Like a lobster shell, security has layers — review code before you run it.
latestvk97agrc8m3a99jx258hwd84z2983nq22
License
MIT-0
Free to use, modify, and redistribute. No attribution required.