Last updated: August 15, 2026
SciREPL is designed to be a privacy-respecting application. All code execution happens locally on your device. We do not collect, store, or transmit any personal data to our servers.
SciREPL stores data locally on your device using browser or app storage, including localStorage, IndexedDB, and caches:
SciREPL does not send this local data to its servers. You can clear session and VFS data via Menu > Clear History, downloaded runtime and app caches via Menu > Memory & Storage, and verified catalogue snapshots, cached workbook artifact bytes, and trusted release mappings via Browse > Catalog source > Clear verified data. Clear verified data does not remove installed catalogue notebook copies; delete those notebooks individually or use Menu > Clear History.
SciREPL bundles its core interface and most language runtimes directly in the app. The following items describe what is bundled and what may be downloaded from third-party services:
webr.r-wasm.org, downloaded when R is first used and cached locally.cdn.jsdelivr.net or unpkg.com, downloaded when Lua is first used and cached locally.docx@9.6.0 from cdn.jsdelivr.net, downloaded when DOCX export is first used.s243a.github.io) and immutable catalogue or workbook bytes from raw.githubusercontent.com. These requests do not intentionally include notebook content, search text, spoken-language choices, fallback preferences, or kernel filters. Selecting a development branch additionally asks api.github.com to resolve it to an exact commit. Other packages and user-requested data may be fetched from package repositories or URLs when you request them, for example with install.packages(), %pip, or file-fetching commands.ko-fi.com. No Ko-fi request is made unless you choose to open it.When these resources are fetched, the CDN providers may receive:
This is standard HTTP behavior and is not under our control. Caching minimizes repeat connections. After a runtime or package is downloaded, SciREPL uses a local cached copy when available. Verified catalogue snapshots, downloaded catalogue workbook artifact bytes, and trusted release mappings are also stored locally; clear those from Browse > Catalog source > Clear verified data. This does not remove installed catalogue notebook copies; delete them individually or use Menu > Clear History. Built-in catalogue items remain available.
jsDelivr's privacy policy is available at jsdelivr.com/terms/privacy-policy-jsdelivr-net.
Your code, session data, and exported files are not transmitted to a SciREPL server. Network requests occur only for the resources described above or when your code explicitly requests them.
After you accept the current network disclosure, opening Browse can contact the configured static catalogue host and raw.githubusercontent.com to load verified release metadata and catalogue content. A development branch additionally uses api.github.com only to resolve that branch to an exact commit. Installing an optional catalogue item downloads its selected workbook or package bytes. These requests do not intentionally include notebook content, search text, spoken-language choices, fallback preferences, or kernel filters. The services receive your IP address and standard HTTP headers. SciREPL does not proxy, log, or monitor these requests. Other package downloads are user-initiated and optional.
SciREPL provides a file-fetching feature that allows notebook code to download files from URLs you specify. When you use this feature (e.g., fetch_file in Prolog, or via the Prolog Settings panel), the request is made directly from your device to the specified URL. The remote server will receive your IP address and standard HTTP headers. SciREPL does not proxy, log, or monitor these requests.
Code you enter in the included Python, Prolog, Bash, JavaScript, R, Lua, TypR, and ClojureScript kernels is executed locally on your device. No code is sent to a SciREPL server for execution.
When you export a notebook (.ipynb), the file is created locally on your device. If you choose to share it (via the system share sheet), the destination is determined by your choice (e.g. local storage, email, cloud services). SciREPL does not control or monitor where you share exported files.
SciREPL does not include any analytics, crash reporting, advertising SDKs, or tracking mechanisms.
SciREPL does not knowingly collect any information from children. The app does not require account creation or collect personal information of any kind.
If a future release changes which runtimes or packages are bundled or downloaded, this policy will be updated to describe the resulting network requests and metadata exposure.
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 PERFORMANCE THEREOF.
You use SciREPL entirely at your own risk. Code execution occurs locally on your device and the developers are not responsible for any consequences of code you choose to run.
If you have questions about this privacy policy, please open an issue on our GitHub repository.