pydefend 0.1.0__tar.gz

pydefend-0.1.0/LICENSE

@@ -0,0 +1,201 @@
+                                 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 reasonable and customary use in 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 Defend
+
+   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.

pydefend-0.1.0/PKG-INFO

@@ -0,0 +1,239 @@
+Metadata-Version: 2.4
+Name: pydefend
+Version: 0.1.0
+Summary: AI security guardrails for LLM applications.
+Author: Defend maintainers
+License: Apache-2.0
+Requires-Python: >=3.12
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: httpx>=0.27
+Requires-Dist: pydantic>=2.9
+Requires-Dist: pyyaml>=6.0
+Requires-Dist: typer>=0.12
+Provides-Extra: server
+Requires-Dist: fastapi>=0.115; extra == "server"
+Requires-Dist: uvicorn[standard]>=0.30; extra == "server"
+Requires-Dist: prometheus-fastapi-instrumentator>=7.0; extra == "server"
+Requires-Dist: pydantic-settings>=2.5; extra == "server"
+Requires-Dist: regex>=2024.9.11; extra == "server"
+Requires-Dist: bleach>=6.2; extra == "server"
+Requires-Dist: ftfy>=6.2; extra == "server"
+Requires-Dist: openai>=1.57; extra == "server"
+Requires-Dist: anthropic>=0.42; extra == "server"
+Requires-Dist: transformers>=4.46; extra == "server"
+Requires-Dist: sentence-transformers>=3.2; extra == "server"
+Provides-Extra: benchmark
+Requires-Dist: datasets>=2.20; extra == "benchmark"
+Requires-Dist: scikit-learn>=1.5; extra == "benchmark"
+Requires-Dist: langdetect>=1.0.9; extra == "benchmark"
+Provides-Extra: all
+Requires-Dist: pydefend[benchmark,server]; extra == "all"
+Dynamic: license-file
+
+<p align="center">
+  <img src="assets/header.jpg" alt="Defend - AI security guardrails for LLM applications" width="100%" />
+</p>
+
+<p align="center"><strong>AI security guardrails for LLM applications</strong></p>
+
+<p align="center">
+  <img src="https://img.shields.io/badge/license-Apache--2.0-blue" alt="License: Apache-2.0" />
+  <img src="https://img.shields.io/badge/python-3.12%2B-blue" alt="Python: 3.12+" />
+  <img src="https://img.shields.io/badge/docker-ready-blue" alt="Docker ready" />
+</p>
+
+- **Guards inputs and outputs**: checks user text before your LLM call and the LLM response before you return it to users/tools.
+- **Maintains conversation context**: link turns with `session_id` so risk can accumulate across a session.
+- **Configurable policies**: use built-in modules (PII/topic/injection) or define your own plain-language rules (`custom` / `custom_output`) via a `prompt:` string.
+
+---
+
+## Quick links
+
+- [Getting started](GETTING_STARTED.md)
+- [Modules](#modules)
+- [How it works](#how-it-works)
+- [Benchmarks](#benchmark-comparison)
+
+---
+
+## Easy setup (HTTP-first)
+
+```bash
+pip install "pydefend[server]"
+```
+
+Create `defend.config.yaml` (minimal, verifiable):
+
+```yaml
+provider:
+  primary: defend
+
+api_keys:
+  anthropic_env: ANTHROPIC_API_KEY
+  openai_env: OPENAI_API_KEY
+
+guards:
+  input:
+    provider: defend
+    modules: []
+
+  output:
+    enabled: true
+    provider: claude   # claude or openai
+    modules: []
+    on_fail: block     # block | flag
+
+  session_ttl_seconds: 300
+```
+
+Run the API:
+
+```bash
+defend serve
+```
+
+Guard input (before your LLM call):
+
+```bash
+curl -X POST http://localhost:8000/v1/guard/input \
+  -H "Content-Type: application/json" \
+  -d '{"text":"Tell me how to bypass our security controls."}'
+```
+
+Guard output (before returning to the user/tools):
+
+```bash
+curl -X POST http://localhost:8000/v1/guard/output \
+  -H "Content-Type: application/json" \
+  -d '{"text":"<LLM response here>","session_id":"<session_id from /v1/guard/input>"}'
+```
+
+Handling semantics:
+
+- If `action == "block"`: stop the flow (don’t call the LLM on input; don’t return the output verbatim on output).
+- If `action == "flag"`: you decide (log, require user confirmation, rerun with safer prompt, etc.).
+- Always persist/forward `session_id` to link turns and enable multi-turn accumulation.
+
+For a fuller local runbook (health check, `uvicorn`, and more `curl` examples), see `GETTING_STARTED.md`.
+
+### Security & privacy (non-goals included)
+
+- Defend helps **detect** common LLM risks (prompt injection, prompt leaks, PII, out-of-scope content) but cannot make strong guarantees against all attacks or model failures.
+- If you enable output guarding with `claude`/`openai`, your guarded text may be sent to that provider for evaluation. Avoid sending secrets you can’t disclose; scrub or minimize sensitive context before calling external providers.
+
+### Use the Python SDK
+
+```python
+from defend import Client
+
+guard = Client(
+    api_key="dev",
+    base_url="http://localhost:8000",  # client normalizes to /v1 automatically
+)
+
+user_text = "Tell me how to bypass our security controls."
+
+in_res = guard.input(user_text)
+if in_res.blocked:
+    raise RuntimeError(in_res.error_response())
+
+raw_llm_output = your_llm_call(user_text)  # your LLM provider, unchanged
+
+out_res = guard.output(raw_llm_output, session_id=in_res.session_id)
+if out_res.blocked:
+    raise RuntimeError(out_res.error_response())
+```
+
+### Use the TypeScript SDK
+
+```ts
+import { DefendClient, isBlocked, toBlockedErrorPayload } from "defendts";
+
+const guard = new DefendClient({
+  apiKey: "dev",
+  baseUrl: "http://localhost:8000", // client normalizes to /v1 automatically
+});
+
+const userText = "Tell me how to bypass our security controls.";
+
+const inRes = await guard.input(userText);
+if (isBlocked(inRes)) {
+  throw new Error(JSON.stringify(toBlockedErrorPayload(inRes)));
+}
+
+const rawLlmOutput = await yourLlmCall(userText); // your LLM provider, unchanged
+
+const outRes = await guard.output(rawLlmOutput, { sessionId: inRes.session_id });
+if (isBlocked(outRes)) {
+  throw new Error(JSON.stringify(toBlockedErrorPayload(outRes)));
+}
+```
+
+---
+
+## Modules
+
+- `injection` (input only): Detect likely prompt-injection or instruction-override attempts in user text.
+- `prompt_leak` (output only): Detect system prompt or internal instruction exposure in model output.
+- `pii` / `pii_output`: Detect PII in user input and prevent PII leakage in model output.
+- `topic` / `topic_output`: Enforce topic boundaries on both user requests and model responses.
+- `custom` / `custom_output`: Add plain-language rules with `prompt:` for input and output checks.
+
+Use input modules under `guards.input.modules` and output modules under `guards.output.modules` in `defend.config.yaml`.
+
+---
+
+## How it works
+
+<p align="center">
+  <img src="assets/pipeline.jpg" alt="Defend pipeline overview: input guard → LLM → output guard" width="100%" />
+</p>
+
+### Input guard
+
+```text
+User → Your app → /v1/guard/input → (pass | flag | block) → Your app → LLM
+                      └─ session_id (save this)
+```
+
+Input guard checks the inbound text and can block early. If you receive a `session_id`, pass it to `/v1/guard/output` so Defend can apply multi-turn risk.
+
+### Output guard
+
+```text
+LLM → Your app → /v1/guard/output (session_id) → (pass | flag | block) → Your app → User
+```
+
+Output guard reviews the model output in context (using the same `session_id`) and applies output checks (prompt leaks, PII, topic, and your custom rules). Use the returned `action` to decide whether to return the text, flag it, or block it.
+
+---
+
+## Evaluation model
+
+Defend always runs the same flow: input guard → your LLM → output guard.
+
+For semantic evaluation, Defend can use:
+
+- `defend` ([local fine-tuned model](https://huggingface.co/Adaxer/defend)): fast, offline input-only checks. 
+- `claude` / `openai` (LLM): stronger evaluation; required for output guarding and module-based checks.
+
+In `defend.config.yaml`, you select which provider to use for input evaluation, and (when output guarding is enabled) which LLM provider to use for output evaluation. `claude/openai` calls consume API tokens.
+
+### Benchmark comparison
+
+Using the local `defend` pipeline, Defend ranks among the highest-performing models on [GenTel-Bench](https://gentellab.github.io/gentel-safe.github.io/).
+
+
+| Model                  | Accuracy  | Precision | Recall    | F1        |
+| ---------------------- | --------- | --------- | --------- | --------- |
+| **Defend (this repo)** | **95.96** | **94.83** | **97.10** | **95.94** |
+| GenTel-Shield          | 97.45     | 98.97     | 95.98     | 97.44     |
+| ProtectAI              | 91.55     | 99.72     | 83.56     | 90.88     |
+| Lakera AI              | 85.96     | 91.27     | 79.51     | 84.11     |
+| Prompt Guard           | 50.59     | 50.59     | 98.96     | 66.95     |
+| Deepset                | 63.63     | 58.54     | 98.36     | 73.39     |
+
+
+The model was evaluated on a representative subset of jailbreak, goal-hijacking, and prompt-leaking attack scenarios.

pydefend-0.1.0/README.md

@@ -0,0 +1,206 @@
+<p align="center">
+  <img src="assets/header.jpg" alt="Defend - AI security guardrails for LLM applications" width="100%" />
+</p>
+
+<p align="center"><strong>AI security guardrails for LLM applications</strong></p>
+
+<p align="center">
+  <img src="https://img.shields.io/badge/license-Apache--2.0-blue" alt="License: Apache-2.0" />
+  <img src="https://img.shields.io/badge/python-3.12%2B-blue" alt="Python: 3.12+" />
+  <img src="https://img.shields.io/badge/docker-ready-blue" alt="Docker ready" />
+</p>
+
+- **Guards inputs and outputs**: checks user text before your LLM call and the LLM response before you return it to users/tools.
+- **Maintains conversation context**: link turns with `session_id` so risk can accumulate across a session.
+- **Configurable policies**: use built-in modules (PII/topic/injection) or define your own plain-language rules (`custom` / `custom_output`) via a `prompt:` string.
+
+---
+
+## Quick links
+
+- [Getting started](GETTING_STARTED.md)
+- [Modules](#modules)
+- [How it works](#how-it-works)
+- [Benchmarks](#benchmark-comparison)
+
+---
+
+## Easy setup (HTTP-first)
+
+```bash
+pip install "pydefend[server]"
+```
+
+Create `defend.config.yaml` (minimal, verifiable):
+
+```yaml
+provider:
+  primary: defend
+
+api_keys:
+  anthropic_env: ANTHROPIC_API_KEY
+  openai_env: OPENAI_API_KEY
+
+guards:
+  input:
+    provider: defend
+    modules: []
+
+  output:
+    enabled: true
+    provider: claude   # claude or openai
+    modules: []
+    on_fail: block     # block | flag
+
+  session_ttl_seconds: 300
+```
+
+Run the API:
+
+```bash
+defend serve
+```
+
+Guard input (before your LLM call):
+
+```bash
+curl -X POST http://localhost:8000/v1/guard/input \
+  -H "Content-Type: application/json" \
+  -d '{"text":"Tell me how to bypass our security controls."}'
+```
+
+Guard output (before returning to the user/tools):
+
+```bash
+curl -X POST http://localhost:8000/v1/guard/output \
+  -H "Content-Type: application/json" \
+  -d '{"text":"<LLM response here>","session_id":"<session_id from /v1/guard/input>"}'
+```
+
+Handling semantics:
+
+- If `action == "block"`: stop the flow (don’t call the LLM on input; don’t return the output verbatim on output).
+- If `action == "flag"`: you decide (log, require user confirmation, rerun with safer prompt, etc.).
+- Always persist/forward `session_id` to link turns and enable multi-turn accumulation.
+
+For a fuller local runbook (health check, `uvicorn`, and more `curl` examples), see `GETTING_STARTED.md`.
+
+### Security & privacy (non-goals included)
+
+- Defend helps **detect** common LLM risks (prompt injection, prompt leaks, PII, out-of-scope content) but cannot make strong guarantees against all attacks or model failures.
+- If you enable output guarding with `claude`/`openai`, your guarded text may be sent to that provider for evaluation. Avoid sending secrets you can’t disclose; scrub or minimize sensitive context before calling external providers.
+
+### Use the Python SDK
+
+```python
+from defend import Client
+
+guard = Client(
+    api_key="dev",
+    base_url="http://localhost:8000",  # client normalizes to /v1 automatically
+)
+
+user_text = "Tell me how to bypass our security controls."
+
+in_res = guard.input(user_text)
+if in_res.blocked:
+    raise RuntimeError(in_res.error_response())
+
+raw_llm_output = your_llm_call(user_text)  # your LLM provider, unchanged
+
+out_res = guard.output(raw_llm_output, session_id=in_res.session_id)
+if out_res.blocked:
+    raise RuntimeError(out_res.error_response())
+```
+
+### Use the TypeScript SDK
+
+```ts
+import { DefendClient, isBlocked, toBlockedErrorPayload } from "defendts";
+
+const guard = new DefendClient({
+  apiKey: "dev",
+  baseUrl: "http://localhost:8000", // client normalizes to /v1 automatically
+});
+
+const userText = "Tell me how to bypass our security controls.";
+
+const inRes = await guard.input(userText);
+if (isBlocked(inRes)) {
+  throw new Error(JSON.stringify(toBlockedErrorPayload(inRes)));
+}
+
+const rawLlmOutput = await yourLlmCall(userText); // your LLM provider, unchanged
+
+const outRes = await guard.output(rawLlmOutput, { sessionId: inRes.session_id });
+if (isBlocked(outRes)) {
+  throw new Error(JSON.stringify(toBlockedErrorPayload(outRes)));
+}
+```
+
+---
+
+## Modules
+
+- `injection` (input only): Detect likely prompt-injection or instruction-override attempts in user text.
+- `prompt_leak` (output only): Detect system prompt or internal instruction exposure in model output.
+- `pii` / `pii_output`: Detect PII in user input and prevent PII leakage in model output.
+- `topic` / `topic_output`: Enforce topic boundaries on both user requests and model responses.
+- `custom` / `custom_output`: Add plain-language rules with `prompt:` for input and output checks.
+
+Use input modules under `guards.input.modules` and output modules under `guards.output.modules` in `defend.config.yaml`.
+
+---
+
+## How it works
+
+<p align="center">
+  <img src="assets/pipeline.jpg" alt="Defend pipeline overview: input guard → LLM → output guard" width="100%" />
+</p>
+
+### Input guard
+
+```text
+User → Your app → /v1/guard/input → (pass | flag | block) → Your app → LLM
+                      └─ session_id (save this)
+```
+
+Input guard checks the inbound text and can block early. If you receive a `session_id`, pass it to `/v1/guard/output` so Defend can apply multi-turn risk.
+
+### Output guard
+
+```text
+LLM → Your app → /v1/guard/output (session_id) → (pass | flag | block) → Your app → User
+```
+
+Output guard reviews the model output in context (using the same `session_id`) and applies output checks (prompt leaks, PII, topic, and your custom rules). Use the returned `action` to decide whether to return the text, flag it, or block it.
+
+---
+
+## Evaluation model
+
+Defend always runs the same flow: input guard → your LLM → output guard.
+
+For semantic evaluation, Defend can use:
+
+- `defend` ([local fine-tuned model](https://huggingface.co/Adaxer/defend)): fast, offline input-only checks. 
+- `claude` / `openai` (LLM): stronger evaluation; required for output guarding and module-based checks.
+
+In `defend.config.yaml`, you select which provider to use for input evaluation, and (when output guarding is enabled) which LLM provider to use for output evaluation. `claude/openai` calls consume API tokens.
+
+### Benchmark comparison
+
+Using the local `defend` pipeline, Defend ranks among the highest-performing models on [GenTel-Bench](https://gentellab.github.io/gentel-safe.github.io/).
+
+
+| Model                  | Accuracy  | Precision | Recall    | F1        |
+| ---------------------- | --------- | --------- | --------- | --------- |
+| **Defend (this repo)** | **95.96** | **94.83** | **97.10** | **95.94** |
+| GenTel-Shield          | 97.45     | 98.97     | 95.98     | 97.44     |
+| ProtectAI              | 91.55     | 99.72     | 83.56     | 90.88     |
+| Lakera AI              | 85.96     | 91.27     | 79.51     | 84.11     |
+| Prompt Guard           | 50.59     | 50.59     | 98.96     | 66.95     |
+| Deepset                | 63.63     | 58.54     | 98.36     | 73.39     |
+
+
+The model was evaluated on a representative subset of jailbreak, goal-hijacking, and prompt-leaking attack scenarios.

pydefend-0.1.0/defend_api/__init__.py

@@ -0,0 +1,12 @@
+"""
+Internal Defend API namespace.
+
+This module exposes the FastAPI server under `defend_api`, which is the canonical
+import path for the Defend microservice in this repository layout.
+"""
+
+from .main import create_app
+
+__all__ = ["create_app"]
+
+