Skip writer's block using an agentic plugin to your favorite content environment.
PseudoScribe is an AI-powered writing assistant with a focus on multi-tenant support, containerization, and modular architecture. It helps users overcome writer's block by analyzing text style, providing suggestions, and adapting content to match specific writing styles.
- Style Analysis: Analyze text for complexity, formality, tone, and readability
- Style Comparison: Compare two text samples for stylistic similarity
- Style Adaptation: Adapt text to match a target writing style
- Style Checking: Verify if text matches a target style profile
- Multi-tenant Support: Secure isolation between different users/organizations
- Containerized Deployment: Easy setup with Podman (or Docker) and podman-compose
- Podman (or Docker) and podman-compose
- Git
# Clone the repository
git clone https://github.com/quaid/pseudoscribe.git
cd pseudoscribe
# Start the containerized environment
./scripts/operate_first.py start
As an alternative to Podman, you can run the development environment on Kubernetes using the provided manifests.
Prerequisites:
- A running Kubernetes cluster (e.g., Rancher Desktop, kind, Minikube)
kubectl
configured to connect to your cluster
Installation & Testing:
# From the pseudoscribe directory, run the Rancher test script
./scripts/test-rancher.sh
This script will deploy all services to your Kubernetes cluster, run the test suites, and clean up the resources upon completion.
POST /api/style/analyze
: Analyze text style characteristicsPOST /api/style/compare
: Compare styles of two text samplesPOST /api/style/adapt
: Adapt text to match target stylePOST /api/style/check
: Check text against target style
GET /api/tenant/config
: Get tenant configurationPOST /api/tenant/config
: Update tenant configuration
GET /api/role
: List available rolesPOST /api/role
: Create a new role
PseudoScribe follows the Writer's Block Assistant Development Standards with a focus on:
- Test-Driven Development (TDD)
- "No Mocks or Stubs" policy for production code
- Containerized development workflow
- Behavior-Driven Development (BDD) for testing
- Start with a GitHub Issue
- Create a feature branch (
feature/AI-{id}
,bug/B-{id}
, orchore/CH-{id}
) - Follow TDD workflow (Red Tests → Green Tests → Refactor)
- Create a Pull Request
- Review and merge
PseudoScribe uses a modular architecture with the following components:
- API Layer: FastAPI endpoints
- Infrastructure Layer: Core services and utilities
- Model Layer: Data models and database interactions
- Local-first architecture
- Multi-tenant isolation
- Proper error handling
- Secure API design
This project is licensed under the MIT License - see the LICENSE file for details.