fault_adaptive_spacecraft_docking

Task goal

The ALE engine drops an autonomous agent into a sandboxed robotics problem built from the task's own image, with the network blocked. The agent reads instruction.md — the only thing it is told — develops its solution, and leaves the artifact under /home/user/submission. Scoring happens afterwards: the task's verify stage, absent from the sandbox while the agent works, runs the artifact on hidden seeds and computes the authoritative Mean docking quality. Nothing the agent prints is ever a scoring input.

Scoring

The verifier compares the measured Mean docking quality (higher is better) against the task's anchor — the value the reference implementation in oracle/ actually produced through the same grader, never a figure copied from a paper.

ratio  = clamp(measured / anchor, 0, 1.5)
reward = clamp(ratio / 0.75, 0, 1)

A ratio of 1.0 means the agent matched the reference; beating it gives a ratio above 1.0, capped at 1.5. The reward is the task's score: it rises with the ratio and saturates at 1.0 once the ratio reaches 0.75.

Anchor policy. Everything the grader needs, the anchor included, travels inside the task folder under verify/ and is hidden from the agent by stage isolation — the verify stage is simply absent while the agent works — not by keeping the number secret. That is why the anchor is published here.