Late Night Companion
v1.0.0Proactively offers gentle, non-judgmental emotional support during local late-night hours without forced positivity or unsolicited advice.
⭐ 0· 68·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
Name/description (proactive late-night emotional support) match the SKILL.md: the doc describes auto-trigger conditions, message templates, modes, optional memory fields, and safety guidance. The skill does not request unrelated credentials or binaries. One small inconsistency: the SKILL.md front-matter includes "disabled: true" while the doc describes automatic activation behavior—this likely means the skill is intended to be opt-in by default on the platform, not forcibly enabled.
Instruction Scope
Instructions stay within scope: they describe when to send a single opening message, how to listen/respond, closing behavior, optional local memory fields, and safety escalation. They reference platform signals (local time, session online/activity, do_not_disturb) which are reasonable for proactive behavior. The rules leave some discretionary language (e.g., "MAY send one opening message"), which grants the agent judgement but is within the feature's purpose.
Install Mechanism
No install spec and no code files — instruction-only. This is low-risk: nothing is downloaded or written to disk by the skill itself.
Credentials
The skill requires no environment variables, credentials, or external config paths. The optional memory fields are minimal and clearly scoped to the conversation. No disproportionate secrets or unrelated service tokens are requested.
Persistence & Privilege
The skill is not marked always:true and does not request special privileges, but it is designed to be autonomously invoked during quiet hours (the platform default allows autonomous invocation). That behavior is expected for a proactive companion, but users should be aware it will monitor session/activity/DND signals to decide when to send messages.
Assessment
This skill appears to do what it says: proactively offer one gentle late-night check-in and then listen. Before enabling, verify (1) whether the skill is actually disabled by default (the SKILL.md contains "disabled: true") and how your platform handles enabling it; (2) that the platform's DND/online/activity signals are respected so you won't get unwanted pings; (3) that any optional conversation memory is stored according to your privacy preferences; and (4) you understand the safety note — this is not crisis care and should direct users to local emergency services when needed. If you want stricter limits, require explicit user opt-in rather than relying on automatic triggers.Like a lobster shell, security has layers — review code before you run it.
latestvk9749gwsjv1v3ambjgqznw0yth83kvsk
License
MIT-0
Free to use, modify, and redistribute. No attribution required.