Movement bfs frontier
Skill secondorderai/ouroboros/benchmarks/arc-agi-3/kaggle/skills/movement-bfs-frontier
Ouroboros is an Agent Harness, CLI and Desktop App with Self-Improvement layer
npx -y skills add secondorderai/ouroboros --skill movement-bfs-frontierAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Use inferred direction deltas to search toward unvisited safe positions after simple action probes identify a controlled object.
The file declares its own license as Apache-2.0. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
1.0 KB, as published. Nobody here has run it
Movement BFS Frontier
Use this skill when simple action probes show that actions 1-4 move a stable foreground object. Treat the moved object as the controlled object only after the transition model has observed consistent deltas.
Plan by building a local movement graph from the current inferred position,
learned action deltas, blocked edges, death edges, and visited positions. Prefer
short paths to unvisited safe cells. Never choose an edge already observed to
cause GAME_OVER or no visible change from the same state.
This skill is intentionally generic. It must not store exact game ids, frame hashes, solved-level walkthroughs, or fixed coordinates from public games.