Active Plugins¶

The plugins currently shipped with the repo. The exact set running in any given environment is whatever VaultSettings.json Base.active_plugins says — this page describes what's available to enable.

Note

Endpoint paths below are written without the /vault/ prefix, which is added automatically by NimbusVault/urls.py. So llm/document resolves to https://<host>/vault/llm/document.

LLM_Plugin¶

Purpose: LLM-driven document generation, search, chat, regulatory-guideline assessment, and text-to-* utilities (text-to-SQL, text-to-filter, text-to-workflow).

Folder: Plugins/LLM_Plugin/

Key sub-modules:

Path	Responsibility
`TextToSQL/`	Natural-language → SQL over tenant data. Includes a schema-sync step and a direct-execute path.
`TextToFilter/`	Natural-language → DRF filter clauses for document search.
`TextToWorkflow/`	Natural-language → workflow definition.
`SuggestFields/`	Field auto-suggest for forms.
`RegulatoryMapping/`	Map regulatory guidelines to entity attributes; produce an assessment.

Endpoint surface (Plugins/LLM_Plugin/urls.py):

POST  llm/document
POST  llm/chat
POST  llm/check
POST  llm/modeltrain
POST  llm/modelchat
GET   llm/model-sessions
POST  llm/v<vrsn_id>/doc-assess
POST  llm/sync-attributes-to-docsearch
POST  llm/text-to-filter
POST  llm/regulatory-guideline/file
POST  llm/regulatory-guideline/assessment

(Plus TextToSQL endpoints; consult Plugins/LLM_Plugin/TextToSQL/Views.py for the full list.)

Configuration: LLM provider credentials, model names, and feature toggles live under the LLM block in VaultSettings.json.

Nimbus_Plugin¶

Purpose: Integration glue for the Nimbus product — model integration, pipeline orchestration, and permission/redirection helpers.

Folder: Plugins/Nimbus_Plugin/

Endpoint surface (Plugins/Nimbus_Plugin/urls.py):

POST  nimbus/model-entity/enablemodelintegration
GET   nimbus/model-entity/getnimbusprojectlist
GET   nimbus/getallmodels
GET   nimbus/getpipelinelist
POST  nimbus/runpipeline
GET   nimbus/getpipelinestatus
GET   nimbus/get-user-permission
POST  nimbus/redirection
POST  nimbus/unlink_model

Configuration: Nimbus host, auth, and pipeline routing under the Nimbus block in VaultSettings.json.

SharePoint¶

Purpose: Import / sync documents from Microsoft SharePoint into NimbusVault.

Folder: Plugins/SharePoint/

Endpoint surface (Plugins/SharePoint/urls.py):

GET   sharepoint/files
POST  sharepoint/sync_document
POST  sharepoint/search-files
GET   sharepoint/get-processes
POST  sharepoint/pipeline
GET   sharepoint/get-process-files
POST  sharepoint/stop_search
POST  sharepoint/update-migration

Notes: - Long-running sync work is delegated to Celery. The HTTP endpoints kick off jobs and surface progress via the get-processes / get-process-files reads. - Auth is OAuth2 via the SharePoint app registration; credentials in VaultSettings.json under SharePoint.

PowerBI¶

Purpose: Power BI report metadata + dataset/model lineage exposure; ships a downloadable Power BI connector.

Folder: Plugins/PowerBI/

Endpoint surface (Plugins/PowerBI/urls.py):

GET   PowerBI/Templates
GET   PowerBI/Models
GET   PowerBI/Lineages
GET   PowerBI/download-connector

Configuration: Tenant ID, workspace, and OAuth scope in VaultSettings.json under PowerBI.

TeamsPlugin¶

Purpose: Propagate model/team membership into NimbusVault artifacts so Teams group changes flow into the entity permission model.

Folder: Plugins/TeamsPlugin/

Endpoint surface (Plugins/TeamsPlugin/urls.py):

POST  teams/propagate_model_team_to_artifact

(The view delegates into Plugins/Nimbus_Plugin/views.py::GetNimbusProjectList — keeping the actual implementation co-located with the related Nimbus logic.)

HDFCUAMPlugin¶

Purpose: HDFC-specific User Access Management (UAM) — user lifecycle, segment management, group management requests, parameter change requests, workflow maker/checker, and reporting. The most complex active plugin.

Folder: Plugins/HDFCUAMPlugin/

Internal layout (non-exhaustive):

Plugins/HDFCUAMPlugin/
├── bll/
│   ├── Auth/, ModelEntity/, ModelId/, Reports/, Segments/,
│   ├── UAMParameters/, UserProfile/, Workflow/, iSACRequest/
│   └── (each sub-domain has its own HelperFunctions / GetApis / PostApis / service)
├── cron/, cronjobs/      # supercronic jobs scheduled in the celery container
├── config.py
├── constants.py
├── docs/                 # HDFC-specific deployment / runbook notes
└── (views, serializers, urls)

Endpoint surface (selected — versioned via <str:version> path segment, e.g., v1):

GET   <version>/uam/users
POST  <version>/uam/users
POST  <version>/uam/users/delete
POST  <version>/uam/users/status
GET   <version>/uam/users/details
POST  <version>/uam/users/session
POST  <version>/uam/users/reinstate
GET   <version>/uam/users/report
POST  <version>/uam/request
GET   <version>/uam/request/prefill
GET   <version>/uam/request/list
GET   <version>/uam/report/generate
... (segments, group-management requests, parameter change requests)

See Plugins/HDFCUAMPlugin/urls.py for the complete list.

Notes: - iSAC (HDFC's internal access-control system) integration is encapsulated under bll/iSACRequest/. - The plugin runs its own cron jobs via supercronic — definitions live under Plugins/HDFCUAMPlugin/cron/.

AldermorePlugin¶

Purpose: Aldermore-specific activity-report download endpoint with a customised report shape.

Folder: Plugins/AldermorePlugin/

Endpoint surface (Plugins/AldermorePlugin/urls.py):

GET   activity/reports/download

Adding to this page¶

When you add a new plugin, append a section here following the template above:

One-line purpose.
Folder path.
Key sub-modules / internal layout (if non-trivial).
Endpoint surface — copy from the plugin's urls.py.
Configuration block name in VaultSettings.json.
Anything non-obvious (long-running jobs, Celery queues, external auth, cron).

The development guide covers the mechanics; this page is the catalogue.