Generate language descriptors and system prompts for Langium DSL evaluation suites.
npx skills add https://github.com/eclipse-langium/langium-ai --skill lai
Please help me install this Agent Skill. Skill: lai Source: https://github.com/eclipse-langium/langium-ai/tree/main/skills/lai Command: npx skills add https://github.com/eclipse-langium/langium-ai --skill lai
It solves the challenge of building reliable AI-powered tooling for a Langium DSL by turning your language project into a descriptor-driven workflow that supports evaluation and continuous prompt refinement.
Tell an AI assistant: "Guide me through running lai init, generating a language descriptor, generating a system prompt, and running lai evaluate to measure and improve my Langium DSL prompt quality."
High-intent search queries and answers about installing and using this skill.
To generate a system prompt for a Langium DSL, you create a YAML language descriptor from your project structure and then synthesize a DSL-aware system prompt directly from that descriptor to guide LLM behavior during evaluation.
You evaluate LLM outputs against a DSL grammar by running evaluation suites that test generated text against real Langium parsing and validation, comparing run histories and analyzing diagnostics to measure prompt pass rates.
A language descriptor in Langium AI tooling is a YAML representation of your project's grammar, services, and examples used to keep system prompt generation aligned with your specific domain-specific language structure.
Yes, you can automate prompt refinement by regenerating the descriptor and system prompt from scratch after updating grammar or validation rules, then re-running evaluations to confirm improvements increased pass rates and reduced diagnostics.
Yes, a YAML language descriptor is required to run Langium AI evaluation suites because it defines the schema for validation and drives the synthesis of the system prompt that guides the LLM during the evaluation loop.
Langium AI evaluation failures occur when the generated LLM output violates real Langium parsing constraints; you diagnose these failures by analyzing evaluation run diagnostics, refining the YAML descriptor, and regenerating the system prompt.
eclipse-langiumGenerate a self-contained SKILL.md document for a Langium-based DSL.
stopitdanRun the recommendation engine evaluation suite and summarize pass rates, NDCG, and failures.
eclipse-langiumGenerate or refine Langium DSL system prompts from language descriptors.
eclipse-langiumGenerate and refine language.descriptor.yml files for Langium DSL projects.
