> For clean Markdown of any page, append .md to the page URL. > For a complete documentation index, see https://docs.nvidia.com/nemo-helix/v0.8.0/documentation/vulnerability-scanning/llms.txt. > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/nemo-helix/_mcp/server. > Scan and audit large language models for jailbreaks, prompt injection, encoding bypasses, and other safety failures using NeMo garak, powered by garak. > **Warning** > > NVIDIA NeMo garak is released with *early access* availability and is subject to limited support and potential API changes in future releases. NVIDIA NeMo garak audits LLMs by probing them with adversarial prompts and detecting failures such as jailbreaks, prompt injection, encoding bypasses, and unsafe output generation. It is powered by [garak](https://github.com/NVIDIA/garak), NVIDIA's open-source LLM vulnerability scanner, and integrates with NeMo Helix so audits can target any model reachable through the Inference Gateway. **Tutorials** **SDK Resources** --- ## Typical Workflow A typical audit looks like the following: 1. Create an [audit target](/documentation/vulnerability-scanning/targets) for the model you want to test. 2. Create an [audit configuration](/documentation/vulnerability-scanning/configurations) that selects which garak probes and detectors to run, along with reporting settings. 3. [Run the audit](/documentation/vulnerability-scanning/tutorials/run-an-audit-locally) and inspect the resulting JSONL, HTML, and hitlog reports. The plugin exposes both [synchronous and asynchronous](/documentation/vulnerability-scanning/sdk-resources) Python entry points for each step. --- ## Setup Before you can run audits, you need a working NeMo Helix install with the garak plugin enabled and a garak interpreter on disk. * Follow [Setup](/documentation/get-started) to install the platform and start local services. * Install garak in a Python virtual environment. By default the plugin invokes `~/.garak/.venv/bin/python -m garak`; override the interpreter path with `NEMO_GARAK_PYTHON` if you installed it elsewhere. * Configure at least one [Inference Gateway provider](/documentation/models-and-inference) so audits can route requests to the model under test. --- ## Upgrading from NeMo Auditor The plugin was renamed from `auditor` to `garak`. Plan for the following when you upgrade an existing deployment: * **Drain audit jobs first.** Let queued and running audit jobs finish before you upgrade. A job queued under the old name references the `auditor` executor profile, the `nhx-auditor-tasks` image, and the `nemo_auditor` task module, none of which exist after the upgrade, so it cannot be dispatched or resumed. * **Completed jobs stay available.** Previously created jobs remain listed and can be retrieved by name through the new `/apis/garak/` routes. * **Update names.** Use `nemo garak` instead of `nemo auditor`, `client.garak` instead of `client.auditor`, `/apis/garak/` instead of `/apis/auditor/`, and `NEMO_GARAK_PYTHON` instead of `NEMO_AUDITOR_GARAK_PYTHON`. * **Rename config overrides.** Move any `auditor:` configuration block, including `job_executor_profile`, to `garak:`. The old key is ignored. * **Permissions are unchanged.** Existing roles and scopes that use the `auditor` names keep working. --- ## Task Guides #### [Audit Targets](/documentation/vulnerability-scanning/targets) Define the model under test — generator type, model identifier, and inference endpoint. #### [Audit Configurations](/documentation/vulnerability-scanning/configurations) Choose probes, detectors, and reporting settings for the audit. #### [Run an Audit Locally](/documentation/vulnerability-scanning/tutorials/run-an-audit-locally) End-to-end walkthrough: create entities, run the audit in-process, read the report artifacts. #### [SDK Resources](/documentation/vulnerability-scanning/sdk-resources) Reference for the `GarakPluginResource(client)` SDK surface: `configs`, `targets`, and `run()`. ## References #### [Configuration Schema](/documentation/vulnerability-scanning/configurations/schema) Field reference for `AuditConfig` and its `system`, `run`, `plugins`, and `reporting` sub-models. #### [Target Schema](/documentation/vulnerability-scanning/targets/schema) Field reference for `AuditTarget` (`type`, `model`, `options`). #### [Selecting Probes](/documentation/vulnerability-scanning/configurations/selecting-probes) `probe_spec`, `probe_tags`, and `detector_spec` syntax with worked examples. #### [Inference Gateway](/documentation/vulnerability-scanning/targets/inference-gateway) How `nhx_uri_spec` resolves a target's URI through a NeMo Helix provider. > Scan and audit large language models for jailbreaks, prompt injection, encoding bypasses, and other safety failures using NeMo garak, powered by garak. ## Docs - [NeMo garak Tutorials](https://docs.nvidia.com/nemo-helix/v0.8.0/documentation/vulnerability-scanning/tutorials.md): - [Run an Audit Locally](https://docs.nvidia.com/nemo-helix/v0.8.0/documentation/vulnerability-scanning/tutorials/run-an-audit-locally.md): - [NeMo garak NeMo Helix SDK Resources](https://docs.nvidia.com/nemo-helix/v0.8.0/documentation/vulnerability-scanning/sdk-resources.md): - [Manage Audit Configurations](https://docs.nvidia.com/nemo-helix/v0.8.0/documentation/vulnerability-scanning/configurations.md): - [Selecting Probes](https://docs.nvidia.com/nemo-helix/v0.8.0/documentation/vulnerability-scanning/configurations/selecting-probes.md): - [Configuration Schema](https://docs.nvidia.com/nemo-helix/v0.8.0/documentation/vulnerability-scanning/configurations/schema.md): - [Manage Audit Targets](https://docs.nvidia.com/nemo-helix/v0.8.0/documentation/vulnerability-scanning/targets.md): - [Routing Targets Through the Inference Gateway](https://docs.nvidia.com/nemo-helix/v0.8.0/documentation/vulnerability-scanning/targets/inference-gateway.md): - [Target Schema](https://docs.nvidia.com/nemo-helix/v0.8.0/documentation/vulnerability-scanning/targets/schema.md):