jev-lander

Featured

Design and verify Jev's landing missions in OpenHarness's viewer, where Jev is the flight computer throttling a booster down to a soft touchdown.

AI & Automation 957 stars 79 forks Updated today MIT

Install

View on GitHub

Quality Score: 93/100

Stars 20%
99
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
100
Issue Health 10%
50
License 10%
100
Description 5%
100

Skill Content

# Jev Lander landing design Jev Lander runs a live landing: a booster falls out of the sky under gravity and Jev (TypeSafe's System One model) reads the telemetry each tick and sets a throttle — CUT, COAST, HOVER or BURN — to bring it down soft. The agent shapes `lander.json`: the gravity, the fuel budget, and the launch height. ## The loop 1. Update `lander.json` (title, gravity, fuel, altitude, a `style` line that tells Jev a landing strategy). The viewer watches it and Jev adapts live — no restart, no second server. 2. `node "$JEV_DSH/toolchain/check.mjs"` verifies the workspace's `lander.json` is valid. Run it before you call a mission done. 3. Watch the descent. Does Jev burn early and hard enough to keep descent in check, flare low, and touch down soft — or hover forever (dull) or slam in even at easy gravity (bad)? That observation is the finding. ## Reading the landing The viewer shows a booster coming down at night: a flame that scales with the throttle, smoke and dust, landing legs, a velocity arrow, the height full burn needs to stop it, a fuel tank that drains, and Jev's throttle gauge with four segments filled by probability. A flight ends in SOFT LANDING or CRASHED, and the next booster drops on its own. Good missions produce a readable loop: Jev lets it fall, burns to check the fall, eases off low, and lands soft. `gravity` is the difficulty dial: the higher it is, the less the engine can brake and the more fuel the descent costs, until the tan...

Details

Author
autonomous-ai
Repository
autonomous-ai/openharness
Created
1 months ago
Last Updated
today
Language
C
License
MIT

Similar Skills

Semantically similar based on skill content — not just same category