What it does
The workbench runs deterministic SHACL-based assessments of digital passport data models, compares two composed profiles, and measures SHACL-only coverage of use cases. It takes file-based inputs (composition profiles, use cases and alignments) and returns JSON. The same operations (assess, coverage, compare, prioritize, recommend-composition, schema, vocabulary, template, capabilities and summarize) are available through the dpawb CLI and through a stdio MCP server. Built-in vocabularies and templates are exposed through discovery commands.
The primary intended user is an AI agent that chains the analytical steps through the package API, the CLI or the MCP server. Developers can also run the CLI directly. The example tree is organised by task: loading and assessing profiles, structural comparison, reduced and extended use-case comparison, and composition recommendation. Each example works as a step-by-step tutorial for people and as a fixed recipe over explicit files for agents.
Running it means installing the Python package, or using the MCP server from the official MCP Registry or its container image. Test data in fixtures/ is synthetic and used for validation. Live-source example inputs are meant for manual runs.
Limitations and requirements
Editable installs may be blocked in constrained environments, where the documented repo-native execution path applies. The local release check requires the wheel package. Live-source examples are for manual runs and are not part of CI validation.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
Digital Passport Assessment Workbench
Deterministic SHACL-based assessment and comparison for digital passport data models.
v0.1.2As shown on registry.modelcontextprotocol.io when we last read it