Release And Version Change

Issue 91 Edition 2026-04-01 3 min read

Not accepted General

Sources: 1 • Confidence: Medium • Updated: 2026-04-13 03:32

Key takeaways

datasette-enrichments-llm has a newly available 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 datasette-llm 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 datasette-llm models with an "enrichments" purpose.

What specific breaking changes, migrations, or new features (beyond the configuration shift) are included in datasette-enrichments-llm 0.2a0 compared to the prior release?
What is the exact datasette-llm configuration interface used by datasette-enrichments-llm (settings keys, file format, precedence rules), and how does it map to the plugin’s UI/CLI behavior?
How is the "enrichments" purpose enforced (UI filtering only vs. hard enforcement at execution time), and what happens when a workflow references a model without that purpose?
Is there any direct operator/product/investor decision read-through stated in the corpus (e.g., recommended rollout approach, cost controls, provider selection guidance)?

Centralizing model configuration in datasette-llm could reduce fragmentation and increase interoperability across Datasette LLM integrations, improving adoption and easing governance for teams using multiple LLM features.
Purpose-scoped allowlisting for enrichments suggests growing emphasis on controllability and policy enforcement, potentially supporting enterprise readiness by limiting which models are permitted in enrichment workflows.
Alpha release 0.2a0 indicates an active development phase and possible near-term churn for users tracking versions, which could affect upgrade cadence and support burden.

Documentation or changelog detailing additional features, migrations, or breaking changes beyond the configuration shift, clarifying whether this is mainly refactor or user-visible expansion.
Clear specification of datasette-llm configuration interface and precedence, plus how it maps to the plugin UI and CLI, indicating a stable and operator-friendly configuration surface.
Explicit behavior description showing purpose enforcement at execution time with defined failure modes when a workflow references a model lacking the enrichments purpose.

If the enrichments purpose is only a UI filter and workflows can still run any configured model, governance and enterprise-readiness read-through weakens.
If the datasette-llm integration introduces confusing configuration precedence or regressions in existing workflows, centralization benefits and adoption read-through weakens.
If subsequent releases roll back or materially change the configuration approach soon after 0.2a0, it suggests instability and raises upgrade and support risk.