Metadata-Version: 2.4
Name: sablier-flow
Version: 1.0.13
Summary: Synthetic alternative-history generation for backtest overfitting detection
Project-URL: Homepage, https://sablier.ai/flow
Project-URL: Documentation, https://docs.sablier.ai
Author-email: Sablier <team@sablier.it>
License: 
                                         Apache License
                                   Version 2.0, January 2004
                                http://www.apache.org/licenses/
        
           TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
        
           1. Definitions.
        
              "License" shall mean the terms and conditions for use, reproduction,
              and distribution as defined by Sections 1 through 9 of this document.
        
              "Licensor" shall mean the copyright owner or entity authorized by
              the copyright owner that is granting the License.
        
              "Legal Entity" shall mean the union of the acting entity and all
              other entities that control, are controlled by, or are under common
              control with that entity. For the purposes of this definition,
              "control" means (i) the power, direct or indirect, to cause the
              direction or management of such entity, whether by contract or
              otherwise, or (ii) ownership of fifty percent (50%) or more of the
              outstanding shares, or (iii) beneficial ownership of such entity.
        
              "You" (or "Your") shall mean an individual or Legal Entity
              exercising permissions granted by this License.
        
              "Source" form shall mean the preferred form for making modifications,
              including but not limited to software source code, documentation
              source, and configuration files.
        
              "Object" form shall mean any form resulting from mechanical
              transformation or translation of a Source form, including but
              not limited to compiled object code, generated documentation,
              and conversions to other media types.
        
              "Work" shall mean the work of authorship, whether in Source or
              Object form, made available under the License, as indicated by a
              copyright notice that is included in or attached to the work
              (an example is provided in the Appendix below).
        
              "Derivative Works" shall mean any work, whether in Source or Object
              form, that is based on (or derived from) the Work and for which the
              editorial revisions, annotations, elaborations, or other modifications
              represent, as a whole, an original work of authorship. For the purposes
              of this License, Derivative Works shall not include works that remain
              separable from, or merely link (or bind by name) to the interfaces of,
              the Work and Derivative Works thereof.
        
              "Contribution" shall mean any work of authorship, including
              the original version of the Work and any modifications or additions
              to that Work or Derivative Works thereof, that is intentionally
              submitted to Licensor for inclusion in the Work by the copyright owner
              or by an individual or Legal Entity authorized to submit on behalf of
              the copyright owner. For the purposes of this definition, "submitted"
              means any form of electronic, verbal, or written communication sent
              to the Licensor or its representatives, including but not limited to
              communication on electronic mailing lists, source code control systems,
              and issue tracking systems that are managed by, or on behalf of, the
              Licensor for the purpose of discussing and improving the Work, but
              excluding communication that is conspicuously marked or otherwise
              designated in writing by the copyright owner as "Not a Contribution."
        
              "Contributor" shall mean Licensor and any individual or Legal Entity
              on behalf of whom a Contribution has been received by Licensor and
              subsequently incorporated within the Work.
        
           2. Grant of Copyright License. Subject to the terms and conditions of
              this License, each Contributor hereby grants to You a perpetual,
              worldwide, non-exclusive, no-charge, royalty-free, irrevocable
              copyright license to reproduce, prepare Derivative Works of,
              publicly display, publicly perform, sublicense, and distribute the
              Work and such Derivative Works in Source or Object form.
        
           3. Grant of Patent License. Subject to the terms and conditions of
              this License, each Contributor hereby grants to You a perpetual,
              worldwide, non-exclusive, no-charge, royalty-free, irrevocable
              (except as stated in this section) patent license to make, have made,
              use, offer to sell, sell, import, and otherwise transfer the Work,
              where such license applies only to those patent claims licensable
              by such Contributor that are necessarily infringed by their
              Contribution(s) alone or by combination of their Contribution(s)
              with the Work to which such Contribution(s) was submitted. If You
              institute patent litigation against any entity (including a
              cross-claim or counterclaim in a lawsuit) alleging that the Work
              or a Contribution incorporated within the Work constitutes direct
              or contributory patent infringement, then any patent licenses
              granted to You under this License for that Work shall terminate
              as of the date such litigation is filed.
        
           4. Redistribution. You may reproduce and distribute copies of the
              Work or Derivative Works thereof in any medium, with or without
              modifications, and in Source or Object form, provided that You
              meet the following conditions:
        
              (a) You must give any other recipients of the Work or
                  Derivative Works a copy of this License; and
        
              (b) You must cause any modified files to carry prominent notices
                  stating that You changed the files; and
        
              (c) You must retain, in the Source form of any Derivative Works
                  that You distribute, all copyright, patent, trademark, and
                  attribution notices from the Source form of the Work,
                  excluding those notices that do not pertain to any part of
                  the Derivative Works; and
        
              (d) If the Work includes a "NOTICE" text file as part of its
                  distribution, then any Derivative Works that You distribute must
                  include a readable copy of the attribution notices contained
                  within such NOTICE file, excluding those notices that do not
                  pertain to any part of the Derivative Works, in at least one
                  of the following places: within a NOTICE text file distributed
                  as part of the Derivative Works; within the Source form or
                  documentation, if provided along with the Derivative Works; or,
                  within a display generated by the Derivative Works, if and
                  wherever such third-party notices normally appear. The contents
                  of the NOTICE file are for informational purposes only and
                  do not modify the License. You may add Your own attribution
                  notices within Derivative Works that You distribute, alongside
                  or as an addendum to the NOTICE text from the Work, provided
                  that such additional attribution notices cannot be construed
                  as modifying the License.
        
              You may add Your own copyright statement to Your modifications and
              may provide additional or different license terms and conditions
              for use, reproduction, or distribution of Your modifications, or
              for any such Derivative Works as a whole, provided Your use,
              reproduction, and distribution of the Work otherwise complies with
              the conditions stated in this License.
        
           5. Submission of Contributions. Unless You explicitly state otherwise,
              any Contribution intentionally submitted for inclusion in the Work
              by You to the Licensor shall be under the terms and conditions of
              this License, without any additional terms or conditions.
              Notwithstanding the above, nothing herein shall supersede or modify
              the terms of any separate license agreement you may have executed
              with Licensor regarding such Contributions.
        
           6. Trademarks. This License does not grant permission to use the trade
              names, trademarks, service marks, or product names of the Licensor,
              except as required for describing the origin of the Work and
              reproducing the content of the NOTICE file.
        
           7. Disclaimer of Warranty. Unless required by applicable law or
              agreed to in writing, Licensor provides the Work (and each
              Contributor provides its Contributions) on an "AS IS" BASIS,
              WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
              implied, including, without limitation, any warranties or conditions
              of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
              PARTICULAR PURPOSE. You are solely responsible for determining the
              appropriateness of using or redistributing the Work and assume any
              risks associated with Your exercise of permissions under this License.
        
           8. Limitation of Liability. In no event and under no legal theory,
              whether in tort (including negligence), contract, or otherwise,
              unless required by applicable law (such as deliberate and grossly
              negligent acts) or agreed to in writing, shall any Contributor be
              liable to You for damages, including any direct, indirect, special,
              incidental, or consequential damages of any character arising as a
              result of this License or out of the use or inability to use the
              Work (including but not limited to damages for loss of goodwill,
              work stoppage, computer failure or malfunction, or any and all
              other commercial damages or losses), even if such Contributor
              has been advised of the possibility of such damages.
        
           9. Accepting Warranty or Additional Liability. While redistributing
              the Work or Derivative Works thereof, You may choose to offer,
              and charge a fee for, acceptance of support, warranty, indemnity,
              or other liability obligations and/or rights consistent with this
              License. However, in accepting such obligations, You may act only
              on Your own behalf and on Your sole responsibility, not on behalf
              of any other Contributor, and only if You agree to indemnify,
              defend, and hold each Contributor harmless for any liability
              incurred by, or claims asserted against, such Contributor by reason
              of your accepting any such warranty or additional liability.
        
           END OF TERMS AND CONDITIONS
        
           APPENDIX: How to apply the Apache License to your work.
        
              To apply the Apache License to your work, attach the following
              boilerplate notice, with the fields enclosed by brackets "[]"
              replaced with your own identifying information. (Don't include
              the brackets!)  The text should be enclosed in the appropriate
              comment syntax for the file format. We also recommend that a
              file or class name and description of purpose be included on the
              same "printed page" as the copyright notice for easier
              identification within third-party archives.
        
           Copyright 2026 Sablier
        
           Licensed under the Apache License, Version 2.0 (the "License");
           you may not use this file except in compliance with the License.
           You may obtain a copy of the License at
        
               http://www.apache.org/licenses/LICENSE-2.0
        
           Unless required by applicable law or agreed to in writing, software
           distributed under the License is distributed on an "AS IS" BASIS,
           WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
           See the License for the specific language governing permissions and
           limitations under the License.
License-File: LICENSE
Keywords: backtesting,generative-models,overfitting,quantitative-finance,synthetic-data
Classifier: Development Status :: 5 - Production/Stable
Classifier: Intended Audience :: Financial and Insurance Industry
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Office/Business :: Financial :: Investment
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.10
Requires-Dist: cryptography>=42.0
Requires-Dist: httpx>=0.27
Requires-Dist: numpy<2.0,>=1.24
Requires-Dist: pandas<3.0,>=2.0
Requires-Dist: pyarrow>=14.0
Requires-Dist: pydantic<3.0,>=2.0
Requires-Dist: scipy>=1.11
Requires-Dist: tqdm>=4.65
Requires-Dist: typing-extensions>=4.8
Provides-Extra: adapters-backtrader
Requires-Dist: backtrader>=1.9.78; extra == 'adapters-backtrader'
Provides-Extra: adapters-vectorbt
Requires-Dist: vectorbt>=0.26; extra == 'adapters-vectorbt'
Provides-Extra: all
Requires-Dist: backtrader>=1.9.78; extra == 'all'
Requires-Dist: fastapi>=0.110; extra == 'all'
Requires-Dist: google-cloud-tasks>=2.16; extra == 'all'
Requires-Dist: hypothesis>=6.92; extra == 'all'
Requires-Dist: mkdocs-gen-files>=0.5; extra == 'all'
Requires-Dist: mkdocs-material>=9.5; extra == 'all'
Requires-Dist: mkdocstrings[python]>=0.25; extra == 'all'
Requires-Dist: mypy>=1.10; extra == 'all'
Requires-Dist: pytest-asyncio>=0.23; extra == 'all'
Requires-Dist: pytest-cov>=4.1; extra == 'all'
Requires-Dist: pytest>=8.0; extra == 'all'
Requires-Dist: ruff>=0.4.0; extra == 'all'
Requires-Dist: scikit-learn>=1.3; extra == 'all'
Requires-Dist: scipy>=1.11; extra == 'all'
Requires-Dist: sqlalchemy>=2.0; extra == 'all'
Requires-Dist: vectorbt>=0.26; extra == 'all'
Provides-Extra: dev
Requires-Dist: fastapi>=0.110; extra == 'dev'
Requires-Dist: google-cloud-tasks>=2.16; extra == 'dev'
Requires-Dist: hypothesis>=6.92; extra == 'dev'
Requires-Dist: mkdocs-gen-files>=0.5; extra == 'dev'
Requires-Dist: mkdocs-material>=9.5; extra == 'dev'
Requires-Dist: mkdocstrings[python]>=0.25; extra == 'dev'
Requires-Dist: mypy>=1.10; extra == 'dev'
Requires-Dist: pytest-asyncio>=0.23; extra == 'dev'
Requires-Dist: pytest-cov>=4.1; extra == 'dev'
Requires-Dist: pytest>=8.0; extra == 'dev'
Requires-Dist: ruff>=0.4.0; extra == 'dev'
Requires-Dist: sqlalchemy>=2.0; extra == 'dev'
Provides-Extra: validation
Requires-Dist: scikit-learn>=1.3; extra == 'validation'
Requires-Dist: scipy>=1.11; extra == 'validation'
Description-Content-Type: text/markdown

# sablier-flow

> **Stop shipping overfit backtests.**
> Run your strategy on **N alternative versions of history** that share your data's statistical fingerprint. If the strategy only works on the one specific past that happened, that's a problem you can now measure.

---

## Canonical 5-line workflow

```python
import sablier_flow as sf
df = your_data  # multivariate time series, DatetimeIndex
fit  = sf.fit(df, features=df.columns.tolist(), data_types={c: 'price' for c in df.columns}, horizon=21)
gen  = sf.generate(fit.model_id, n_paths=100, like=df.iloc[-21:])
synth_results = [my_backtest(d) for d in gen.as_dataframes()]
verdict = sf.robustness(my_backtest(df), synth_results, primary_metric='sharpe')
```

`gen.as_dataframes()` returns `list[pd.DataFrame]`, one per synthetic alternative-history path; your existing `my_backtest(df) -> dict` runs on each unchanged.

## What you get in 30 lines

```python
import pandas as pd
import sablier_flow as sf

# 0. Auth — set SABLIER_FLOW_API_KEY env var, or pass api_key="sk_live_..." per call.
# 1. Load your data — any DataFrame with a DatetimeIndex + numeric columns.
real = pd.read_parquet("my_universe.parquet")
backtest_window = real.loc["2023-01-01":"2024-01-01"]    # the slice you'll evaluate

# 2. Fit (one-time, several minutes). 80/20 train/OOS split + 21-bar embargo by
#    default. The held-out OOS slice is kept encrypted next to the model so
#    sf.validate() picks it up automatically. `data_types=` is required on
#    every call — the bundled demo attaches the canonical map on
#    `df.attrs['data_types']`; for your own data, pass an explicit dict
#    like {'AAPL': 'price', '10Y': 'rate'}.
fit = sf.fit(real, features=list(real.columns),
             data_types=real.attrs['data_types'],
             horizon=252, train_split=0.8, embargo_days=21, seed=42)

# 3. Validate the model on the held-out OOS slice — full structural metric
#    suite (calibration, dependence, tails, dynamics, memorization).
report = sf.validate(fit.model_id, data_types=real.attrs['data_types'])
print(report.overall)                  # 'pass' | 'warn' | 'fail'
print(report.memorization_risk)        # 'low' | 'medium' | 'high'

# 4. Generate N synthetic alternative-history paths shaped like your backtest
#    window. ``like=`` derives length + dates + price anchor from the window.
paths = sf.generate(fit.model_id, n_paths=1000, like=backtest_window,
                    data_types=real.attrs['data_types'], seed=42)
synth_dfs = paths.as_dataframes()      # list[pd.DataFrame], one per path

# 5. Run *your existing* backtest on the real window AND on each synthetic.
real_result   = my_backtest(backtest_window)
synth_results = [my_backtest(df) for df in synth_dfs]

# 6. The smoking gun.
verdict = sf.robustness(real_result, synth_results, primary_metric="sharpe")
print(verdict.verdict)                 # 'robust' | 'borderline' | 'overfit' | 'highly_overfit'
print(verdict.overfit_score)           # 0.85 → real beat 85% of synthetic → overfit
print(verdict.summary())               # one-line English summary you can paste anywhere
verdict.to_html("audit.html")          # shareable single-file report
```

Two-step `fit` + `generate` (instead of one shot) means you fit once and generate as many windows as you want from the same `model_id` — cheap iteration on your strategy without paying to retrain.

See the [getting-started notebook on docs.sablier.ai](https://docs.sablier.ai/notebook/) for the step-by-step walkthrough — it embeds the same code as `examples/00_getting_started.ipynb` and lets you download the raw `.ipynb` to run locally.

---

## Sign up + install

1. **Create an account** at <https://sablier.ai> — email/password or "Sign in with Google" both work. New accounts get free credits to cover several full cycles of the getting-started notebook.
2. **Verify your email** by clicking the confirmation link Sablier sends (Google OAuth users skip this step).
3. **Install the SDK** in a fresh venv:

```bash
python -m venv .venv && source .venv/bin/activate

pip install sablier-flow                           # thin client (~30 MB, no GPU deps)
pip install 'sablier-flow[adapters-backtrader]'    # + backtrader integration
pip install 'sablier-flow[adapters-vectorbt]'      # + vectorbt integration
```

Pin to an exact version (e.g. `sablier-flow==1.0.0`) rather than a range when publishing a backtest audit so the analysis re-runs identically months later. Bump the pin explicitly when you want a newer build.

Transitive deps: `pandas`, `numpy`, `pyarrow`, `httpx`, `cryptography`, `pydantic` — installed automatically with `sablier-flow`. No vendor data libraries (`yfinance` etc.); bundled demo datasets ship inside the wheel.

4. **Authenticate**:

```python
import sablier_flow as sf
sf.login()                       # opens https://sablier.ai/auth/device, click Authorize, done
```

Or pass the API key explicitly (preferred for CI):

```bash
export SABLIER_FLOW_API_KEY=sk_live_<your-token>     # from Dashboard → Settings → API Keys
```

That's the whole setup. The default endpoint `https://flow.sablier.ai/v1` over standard TLS — no `gcloud`, no cert pinning, no extra steps. Credit usage is shown live on the dashboard; a full fit + validate + generate cycle on the bundled demo dataset uses a small fraction of the free starter balance.

---

## What ships

| Capability | API |
|---|---|
| Fit a flow model on your history (joint over all your features at daily / weekly / monthly / quarterly cadence; auto-detected from your `DatetimeIndex`. Intraday lands in 1.1.0.) | `sf.fit(df, features=[...], data_types={...}, horizon=..., train_split=..., embargo_days=...)` |
| Generate N synthetic alternative-history paths, anchored at any window | `sf.generate(model_id, n_paths=..., like=window, data_types={...})` |
| Run the full structural validation suite on the held-out OOS slice | `sf.validate(model_id, data_types={...})` — returns `ValidationReport` with `overall`, `memorization_risk`, and ~20 per-metric entries |
| Score a backtest's overfit (real vs synthetic distribution) | `sf.robustness(real_result, synth_results)` — returns `RobustnessReport` with `overfit_score`, `verdict`, and `synthetic_*` percentiles |
| Deflated Sharpe Ratio under two nulls (analytical Bailey-LdP + empirical synthetic-best-of-N) | `sf.deflated_sharpe(...)` or `verdict.deflated_sharpe(n_trials=N)` |
| Evaluate a *family* of strategy variants — CSCV PBO + family-best DSR | `sf.evaluate_family({"name": fn, ...}, real_data, n_paths=...)` |
| Live drift monitoring once a strategy is deployed | `sf.consistency_check(realized_metric, baseline=robustness_report)` |
| List / inspect / delete your fitted models | `sf.list_models()`, `sf.get_model(model_id)`, `sf.delete_model(model_id)` |
| Bundled demo datasets so you can try it with zero data setup | `sf.demo_data()` — daily SPY/QQQ/IWM/TLT + macro series; `sf.demo_data('us_equities_macro_5min_3mo')` — 5-min intraday |
| Engine adapters | `result.as_dataframes()` for pandas; `from sablier_flow.adapters import as_backtrader_feeds, as_vectorbt_panel, write_lean_csv_universe` |

Full API reference: [docs.sablier.ai](https://docs.sablier.ai).

---

## What happens under the hood

```
your laptop ──HTTPS──> Sablier API (Cloud Run) ──Cloud Tasks──> GPU worker (Cloud Run + L4)
     │                                                                    │
     │   1. POST /v1/jobs                                                  │
     │   ◄── 2. ephemeral X25519 pubkey + image digest                     │
     │                                                                     │
     │   3. envelope-encrypt your DataFrame (X25519 + AES-256-GCM)         │
     │   ──> PUT /v1/jobs/{id}/data ──────────────────────────────────────►│
     │                                                                     │
     │                                          4. decrypt in worker RAM,  │
     │                                             train the flow model,   │
     │                                             generate N paths,       │
     │                                             AES-GCM-encrypt back    │
     │                                                                     │
     │   5. GET /v1/jobs/{id}/result ◄────────────────────────────────────-│
     │   6. decrypt locally; result.paths_returns is yours                 │
     ▼
backtester (pandas / backtrader / vectorbt / LEAN / your own)
```

### Security posture today (alpha)

Honest picture of what the SDK actually guarantees right now:

| Layer | Status |
|---|---|
| **TLS 1.3 in transit** (client ↔ API ↔ worker) | ✓ |
| **One-shot AES-256-GCM symmetric key per job** (wrapped in X25519 envelope to the worker's ephemeral pubkey; never re-used, never persisted) | ✓ |
| **GCS at-rest encryption** with Cloud KMS-managed keys (checkpoints + OOS holdouts + result blobs) | ✓ |
| **Customer data isolation** — each job runs in its own Cloud Run instance, scaled to zero between jobs | ✓ |
| **Image-digest pinning** — the SDK ships a pinned digest of the worker image; mismatched server image is rejected before any data is sent | ✓ |
| **AMD SEV-SNP CPU memory encryption** — encrypted RAM, so even a privileged host OS or GCP operator cannot inspect plaintext during training | 🚧 Not yet — Cloud Run L4 is not a confidential VM. Plaintext customer data exists in worker RAM during the ~minutes-long training job. |
| **NVIDIA H100 CC mode** (GPU memory encryption) + **NRAS attestation chain** | 🚧 Awaiting H100 quota |
| **Cryptographic attestation** verified against AMD / NVIDIA root keys before the customer's encryption key is released | 🚧 Same gate — the SDK's `AttestationVerifier` exists and runs the protocol, but the digest pinned today corresponds to a regular Cloud Run image, not a measured-boot enclave |

**What this means concretely**: today the SDK delivers strong network-layer + storage-layer + key-lifecycle protection that's meaningfully better than most quant data SaaS offerings. It does **not** yet deliver memory-encryption-grade protection against a privileged GCP operator. A CISO evaluating us before SEV-SNP + H100 CC ship needs to see and accept that trade-off.

The full SEV-SNP + H100 CC + NRAS attestation deploys with **v0.6**, which lands when GCP releases our H100 confidential-compute quota. The wire protocol the SDK already speaks is the same one we'll use post-rollout, so customer code does not change.

---

## Try it in 20 minutes — no setup beyond `pip install`

```bash
pip install sablier-flow matplotlib
```

Open [`examples/00_getting_started.ipynb`](examples/00_getting_started.ipynb) and run the cells top-to-bottom. The notebook uses the bundled demo dataset (`sf.demo_data()` for daily SPY/QQQ/IWM/TLT + VIX/TNX/DXY macro series; `sf.demo_data('us_equities_macro_5min_3mo')` for 5-min intraday), so it works with zero data setup and zero external network beyond the SDK's hosted endpoint.

---

## Engine integrations

`sablier-flow` is a data layer, not a backtest engine. We integrate with whatever you already use:

| Engine | How |
|---|---|
| Raw pandas / numpy | `result.as_dataframes(index=...)` → `list[pd.DataFrame]`; `from sablier_flow.adapters import as_array` → `np.ndarray` |
| backtrader | `from sablier_flow.adapters import as_backtrader_feeds` → `list[bt.feeds.PandasData]` |
| vectorbt | `from sablier_flow.adapters import as_vectorbt_panel` → wide `pd.DataFrame` |
| LEAN / QuantConnect | `from sablier_flow.adapters import write_lean_csv_universe` → per-path CSV directories |
| In-house C++ / KDB / proprietary | `result.paths_prices` is plain `np.float32[n_paths, horizon, n_features]` — shim ≤ 50 lines |

---

## Why "in-sample is correct" — short version

The model is trained on the same history your backtest will run on. If that triggers your overfit alarm, the answer is in the memorization metric: every fit ships a `memorization_risk` flag computed from the ratio of median synthetic-to-training NN distance over median training-to-training NN distance. If the model is regurgitating samples, the synth lands on top of training points, the ratio collapses below 0.50, the SDK marks it `high`, and the overfit verdict on top of it is explicitly not to be trusted. When it's `low` or `medium`, the synthetic distribution lives in the data-generating process's *neighborhood*, not on top of the training points themselves — which is precisely what you want to stress-test a strategy against. See [`docs/concepts/in-sample-is-correct.md`](docs/concepts/in-sample-is-correct.md) for the long form.

---

## License

`sablier-flow` (this package) — **Apache-2.0**.

---

## Links

- [Getting-started notebook](examples/00_getting_started.ipynb)
- [Concepts docs](docs/concepts/)
- [sablier.ai](https://sablier.ai) — sign up + manage account
- [docs.sablier.ai](https://docs.sablier.ai) — SDK reference + notebook + recipes
- Bug reports / feedback: <team@sablier.it>
