Release And Versioning

Issue 91 Edition 2026-04-01 3 min read

Not accepted General

Sources: 1 • Confidence: Medium • Updated: 2026-04-02 03:46

Key takeaways

datasette-enrichments-llm has a new published release versioned 0.2a0.
datasette-enrichments-llm now uses datasette-llm to configure and manage LLM models.
Model availability for enrichments can be controlled by configuring models with an "enrichments" purpose.

datasette-enrichments-llm now uses datasette-llm to configure and manage LLM models.

Model availability for enrichments can be controlled by configuring models with an "enrichments" purpose.

What are the full changelog details for version 0.2a0 beyond the model-management and purpose-scoping changes?
What configuration format and precedence rules does datasette-llm apply when datasette-enrichments-llm consumes model settings (including defaults and overrides)?
Is the "enrichments" purpose enforced as a hard allowlist at runtime, and what are the failure modes when no eligible models are configured?
Are there backward-compatibility or migration requirements for existing deployments that previously configured models directly in the plugin?
What tests or verification steps demonstrate that enrichments reliably pick up the intended datasette-llm configured models (including the purpose filter) across environments?

Shift to centralized model configuration via datasette-llm could reduce duplicated configuration and make LLM tooling more interoperable across Datasette plugins, potentially increasing adoption of the shared configuration layer.
Purpose-scoped model allowlisting for enrichments suggests a move toward stronger governance and control over model usage, which could make enrichment workflows more acceptable in regulated or policy-driven environments.
An alpha release version implies ongoing iteration and potential breaking changes; near-term user effort may shift toward testing, migration, and validation rather than immediate expansion of usage.

Release notes or changelog for 0.2a0 shows limited, well-documented changes plus clear migration guidance from prior plugin-specific model configuration.
Documentation or tests demonstrate that enrichments reliably discover datasette-llm configured models and correctly enforce the enrichments purpose across environments and deployment modes.
User reports or issues show that purpose-scoping prevents unintended model exposure while providing clear behavior and messaging when no eligible models are configured.

Changelog reveals additional breaking changes or regressions beyond model-management and purpose-scoping, especially around enrichment workflows or configuration compatibility.
Ambiguous configuration precedence in datasette-llm leads to unpredictable model selection or environment-specific behavior when used by datasette-enrichments-llm.
Enrichments purpose enforcement fails open or produces confusing failure modes, causing accidental model availability or blocking enrichments without clear remediation.