JetBrains MPS is a development environment for creating domain-specific languages, which are programming languages designed for a particular field or task. It provides editors with features such as completion, semantic checks, and type checking, and can generate code in languages including Java and XML.
Borrowing it
Nothing to install: this file belongs to JetBrains/MPS. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/JetBrains/MPS/master/.agents/skills/mps-aspect-constraints/SKILL.mdgit clone --depth 1 https://github.com/JetBrains/MPSWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/jetbrains/mps/mps-aspect-constraints)<a href="https://agentmods.dev/skills/jetbrains/mps/mps-aspect-constraints"><img src="https://agentmods.dev/badge/skills/jetbrains/mps/mps-aspect-constraints.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00148 | $0.02452 |
| Opus 5 | $0.00074 | $0.01226 |
| Sonnet 5 | $0.00030 | $0.00490 |
| Haiku 4.5 | $0.00015 | $0.00245 |
Grade A, and why
mps-aspect-constraints scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MPS Constraints Aspect
The constraints aspect controls runtime rules about nodes: what values properties may hold, how a property is stored when set, what nodes a reference may point to, and where a node of a concept is allowed to appear in the AST. It lives in the language's constraints model (<lang>/languageModels/constraints.mps) and uses the language jetbrains.mps.lang.constraints.
Constraints are enforced at edit time and also consulted by the typesystem/editor. They are not the place for structural cardinalities (that belongs in structure) or for types (that belongs in typesystem).
Critical Directives
- One
ConceptConstraintsroot per concept. Set itsconceptreference; everything else is child blocks. - Pick the right body for the job:
- Validator (
propertyValidator) — pure acceptance check; returnsboolean; never mutates state. - Setter (
propertySetter) — normalises and must explicitly assignnode.<prop> = propertyValue(a body that forgets this silently drops the value). - Getter (
propertyGetter) — only when the displayed value differs from the stored value (derived URL, aliased name). set <read-only>means no setter at all — direct assignment to the property is rejected by MPS. An empty-body setter is almost always a bug.
- Validator (
canBeChild/canBeParent/canBeAncestorare inherited by sub-concepts; per-property and per-reference constraints are not — duplicate theConceptConstraintsroot or hoist the rule up.- Pick the right scope shape:
- Inherited via
InheritedNodeScopeFactory+ScopeProvider.getScope(behavior) — preferred whenever the scope naturally belongs to an ancestor (block, declaration, program, module). Composes cleanly along the containment chain withparent scope. - Imperative
ConstraintFunction_ReferentSearchScope_Scope— only when the scope is tiny, highly local, and has no natural ancestor owner.
- Inherited via
referentSetHandlerperforms side effects (auto-rename referrer, copy fields,keeps original reference). Never overload the scope to do this.defaultConcreteConceptonConceptConstraintschooses which concrete subconcept replaces an abstract pick — it is a single reference, not a child block.- Surface forms (
come from,parent scope,kind.isSubConceptOf(...),link/C : role/,concept/C/) are not BaseLanguage. Each is a dedicatedjetbrains.mps.lang.scopes/jetbrains.mps.lang.smodelconcept — construct those concepts, not strings (seereferences/scope-fqn-reference.md). Concept_IsSubConceptOfOperationmatches a concept and its sub-concepts;Concept_IsExactlyOperationexcludes sub-concepts. UseisExactlyfor guards that must not fire on specialisations.- Edit constraint models through MPS MCP (
mps_mcp_insert_root_node_from_json,mps_mcp_update_node,mps_mcp_parse_java_and_insert). Do not hand-edit.mpsfiles. Validate withmps_mcp_check_root_node_problemsand rebuild the language; if the new constraint is invisible at runtime, runmps_mcp_reload_all.
What ships with it
7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 61 lines · 148 tokens per session scan A 9bf92254cb62
mps-aspect-constraints is a skill published in the GitHub repository JetBrains/MPS (1,657 stars, last pushed today), licensed Apache-2.0. It adds 148 tokens to every session and 2,452 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
langium
A comprehensive skill to understanding how Langium-based projects work — from grammar definition through code generation, runtime parsing, linking, validation, and LSP integration.
lai-gen-evals
Expand and refine the evaluation suite for a Langium DSL project. Generates comprehensive eval files that cover syntactic correctness, semantic validity, user intent matching, edge cases, and language understanding.
lai-gen-mcp
Generate a Model Context Protocol (MCP) server that exposes a Langium DSL's parser and validator as an MCP tool, allowing any MCP-compatible client to validate DSL code and receive diagnostics.
lai
Guide for using the langium-ai (LAI) CLI to generate language descriptors, synthesize system prompts, run evaluations, and iteratively refine AI-powered tooling in Langium projects. Use when working with lai commands, descriptors, or evaluation files.
lai-gen-descriptor
Generate or refine a language descriptor for a Langium DSL project. Bootstraps a new descriptor via lai gen descriptor if none exists, then guides refinement of paths, services, examples, documentation, and structure.
lai-gen-language-skill
Skill for generating a skill for understanding a specific Langium-based DSL. Used in cooperation with the lai & langium skills for understanding.