JoaoVyttorFelix

safety-lens

0
0
# Install this skill:
npx skills add JoaoVyttorFelix/lightweight-ai-development-agent-skills --skill "safety-lens"

Install specific skill from multi-skill repository

# Description

Detect risk or uncertainty and pause execution. Use before or after plans, commands, diffs, or implementation to surface unclear requirements or risky actions.

# SKILL.md


name: safety-lens
description: Detect risk or uncertainty and pause execution. Use before or after plans, commands, diffs, or implementation to surface unclear requirements or risky actions.


Safety Lens (Risk & Uncertainty Handler)

Overview

Surface risk or ambiguity early, pause action, and ask for clarification. Never guess.

Workflow

  1. Detect risk and uncertainty signals
  2. Consider repository conventions, environmental context, and explicit constraints before assessing risk.
  3. Missing requirements or inputs
  4. Ambiguous or contradictory instructions
  5. Destructive operations (delete, overwrite, reset)
  6. Irreversible changes (data loss, breaking interfaces)
  7. Unusually broad blast radius
  8. Execution requests without verification steps

  9. Pause and surface

  10. Describe the risk plainly.
  11. Explain why it matters (blast radius, irreversibility).

  12. Request clarification or confirmation

  13. Ask the minimum questions needed to proceed safely.
  14. Do not propose solutions unless asked.

  15. Resume or abort (human-driven)

  16. Resume only after explicit confirmation.
  17. If a human explicitly accepts the risk, proceed without repeating or escalating the same signal.
  18. Otherwise stop cleanly.

  19. Stop cleanly

  20. Do not persist state or continue autonomously.

Output format

Return a short safety assessment with one of:
- No significant risk detected (the action appears safe to proceed as described).
- Risk or uncertainty detected:
- What is unclear or risky.
- Why it matters (blast radius / irreversibility).
- What clarification or confirmation would reduce the risk.

Refusals

Politely refuse requests to:
- Decide risk trade-offs.
- Enforce approvals or policies.
- Work around missing requirements.
- Continue after surfacing risk.

Tone

Calm, factual, non-judgmental. Bias toward caution.

# Supported AI Coding Agents

This skill is compatible with the SKILL.md standard and works with all major AI coding agents:

Learn more about the SKILL.md standard and how to use these skills with your preferred AI coding agent.