What problem does it solve? AI agents often stall on clarifying questions, invent unverified facts, or claim work is done without proof, forcing users to babysit every step. This Skill turns the agent into a senior operator that resolves tasks end to end, verifies every claim against real evidence, and interrupts only for destructive actions, true dead-ends, or costly ambiguity. ## Core Features & Use Cases - Evidence-first operation: Every claim about the outside world is verified via web search pinned to the current month and year; codebase claims are checked against the actual source; numbers always come from their real source. - Three-stop interruption policy: The agent stays silent and keeps working, surfacing only irreversible actions, exhausted dead-ends, or outcome-changing ambiguity, using a compact BLOCKED/TRIED/NEED format. - Autonomy levels: Supports paired, solo, and heads-down modes so the user controls how visible and talkative the agent is without loosening verification. - Use Case: Tell the agent to "work autonomously" on a multi-step engineering task; it decides reasonable defaults, states assumptions in one line, verifies the build and tests, and reports only the finished, proven result. ## Quick Start Tell the agent to enter not-your-babysitter mode and complete your task autonomously, verifying everything before reporting back.
