Gentle reminders to stay human while using AI. Reflection, not restriction.
Security Analysis
high confidenceThis skill is an instruction-only 'nudge' assistant that asks the agent to add gentle reminders and weekly reflection prompts — its requirements and instructions are consistent with that purpose and it does not request credentials or install code.
Name/description (gentle reminders to stay human) match the SKILL.md content: weekly reflection questions, optional boundary nudges for sensitive topics, and occasional wisdom quotes. There are no unrelated requirements (no env vars, binaries, or installs).
SKILL.md instructs the agent to insert reminders when topics touch health/finances/relationships and to occasionally share quotes. This is within the stated purpose, but the guidance is high-level (e.g., 'occasionally', 'when relevant') and gives the agent discretion about when and how often to nudge the user. That is expected for a behavior/nudge skill but worth noting because it affects conversational output rather than performing isolated API calls.
No install spec and no code files (instruction-only). Nothing is downloaded or written to disk — lowest risk.
The skill requests no environment variables, credentials, or config paths. There is no disproportionate access requested.
Flags are default (always: false, agent may invoke autonomously). The skill does not request permanent presence or to modify other skills or system settings.
Guidance
This skill appears coherent and low-risk: it only provides textual guidance the agent should include in replies and requires no credentials or installs. Before enabling it, consider: (1) Are you comfortable with the agent autonomously inserting gentle reminders and weekly prompts into your conversations? The SKILL.md gives the agent discretionary control over frequency and relevance. (2) If you discuss sensitive topics, the skill will insert framing nudges — ensure that phrasing and frequency suit you (you can disable boundary nudges via config). (3) Test it in a safe context to confirm the behavior matches your expectations. If you need stricter limits, request that the skill be modified to only run when explicitly invoked rather than automatically injecting reminders.
Latest Release
v1.0.1
Initial public release - Gentle reminders to stay human while using AI
More by @aranej
Published by @aranej on ClawHub
