Open source JumpStar under MIT

Author: jphilleo

CODE_OF_CONDUCT.md

@@ -0,0 +1,20 @@
+# Code of Conduct
+
+This project should be a welcoming place for people who like games, search,
+benchmarks, tooling, and learning in public.
+
+Please:
+
+- be kind and assume good intent when you can
+- make room for beginners and careful questions
+- criticize ideas, measurements, and code without attacking people
+- disclose uncertainty in benchmark claims
+- give credit for other people's work
+- avoid harassment, insults, threats, and discriminatory language
+
+Maintainers may edit, hide, or remove comments and contributions that make the
+project hostile or unsafe. Repeated or severe violations may result in being
+blocked from participating.
+
+If something feels off, please contact the maintainer privately through GitHub
+or open an issue if the concern is safe to discuss in public.

CONTRIBUTING.md

@@ -0,0 +1,47 @@
+# Contributing
+
+Thanks for taking an interest in JumpStar and CCERL. This project is meant to
+make Chinese Checkers engine work easier to inspect, reproduce, challenge, and
+improve.
+
+Good contributions include:
+
+- bug reports with reproduction steps
+- rules or referee fixes with small test cases
+- benchmark-runner improvements
+- external-engine adapters
+- documentation clarifications
+- careful benchmark results that include commands, versions, seeds, and logs
+
+Before opening a pull request:
+
+1. Build and test locally:
+
+   ```sh
+   cmake -S . -B build -DCCZERO_USE_ACCELERATE=OFF
+   cmake --build build --parallel
+   ctest --test-dir build --output-on-failure
+   ```
+
+2. Keep changes focused. A rules fix, benchmark-data update, and UI polish pass
+   should usually be separate pull requests.
+
+3. Be explicit about benchmark claims. Include the ruleset, position suite,
+   simulation/search settings, commit hash, and whether results are official,
+   provisional, or self-reported.
+
+4. Credit upstream projects and prior work. If you adapt an external engine,
+   document the source repository, license, build settings, and rule differences.
+
+## Benchmark Etiquette
+
+CCERL is intended to make disagreement productive. Stronger engines are welcome.
+So are bug reports that lower JumpStar's rating, expose a rule issue, or show a
+better methodology. Please keep the discussion centered on reproducible evidence.
+
+## Model Files
+
+The repository records model labels and checksums, but does not store `.ccpv`
+model weights in Git history. If you add or update model artifacts, prefer
+GitHub Release assets or another explicit download channel, and document hashes.
+

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Joe Philleo
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

README.md

@@ -1,11 +1,12 @@
 # JumpStar Chinese Checkers Engine
 
 This is the public engine and benchmark repository for JumpStar, a
-self-play-trained Chinese Checkers AI. As of the public launch on **2026-05-23**,
-`JumpStar_60` is, to the best of this project's public evidence, the strongest
-openly documented Chinese Checkers AI engine under the `CCERL-2P10-v2`
-benchmark. That claim is benchmark-scoped: stronger private engines may exist,
-and CCERL is designed so public challengers can test the result.
+self-play-trained Chinese Checkers AI created by **Joe Philleo**. As of the
+public launch on **2026-05-23**, `JumpStar_60` is, to the best of this project's
+public evidence, the strongest openly documented Chinese Checkers AI engine
+under the `CCERL-2P10-v2` benchmark. That claim is benchmark-scoped: stronger
+private engines may exist, and CCERL is designed so public challengers can test
+the result.
 
 This repository intentionally contains only the public research surface:
 
@@ -14,6 +15,7 @@ This repository intentionally contains only the public research surface:
 - the CCERL benchmark rules, protocol, positions, and baseline definitions
 - public leaderboard and direct-match evidence
 - a small benchmark runner/tool subset
+- a lightweight static replay viewer
 
 It does not contain the hosted website, Vercel deployment code, private training
 automation, or internal experiment workspaces.
@@ -98,6 +100,8 @@ Release assets and restored to the paths documented in
 
 ## License
 
-No open-source license has been selected yet. Until a license is added, this
-source is public for inspection and reproducibility but is not broadly licensed
-for reuse or redistribution.
+This repository is open source under the [MIT License](LICENSE).
+
+Model checkpoints are not stored in Git history. If model files are published as
+release assets later, their download page should state whether the same license
+or separate model-artifact terms apply.