Model Context Protocol (MCP) server for CEL (Common Expression Language) expressions with compilation, evaluation, and prompt generation capabilities
Four tools with clear naming (verb_noun pattern: cel_generate_prompt, cel_compile, cel_evaluate, cel_create_environment). Descriptions present and adequate (100-150 chars each). Input schemas visible with type definitions and descriptions for all parameters. However, output schemas are NOT documented, critical gap for LLM planning. Parameters lack enums where appropriate (e.g., no constraint on testCase structure). Error handling guidance absent. Tool composition is sound (single responsibility), but missing pagination/limits guidance and recovery hints.
Compiles a CEL expression. Returns the expression's input and output JSON schemas if it compiles successfully.
Creates a CEL environment configuration from a JSON object.
Evaluates a CEL expression against provided test cases. Returns test case results and coverage.
Generates an LLM authoring prompt explaining the exact variables, functions, and types available in the CEL environment.
Output schemas not documented. LLMs cannot plan downstream calls or extract required fields without knowing what each tool returns.
No error handling guidance. Tool descriptions do not explain failure modes, recovery steps, or what the LLM should do if compilation/evaluation fails.
testCases parameter in cel_evaluate lacks structure constraints. The nested object properties (testCase, bindings, expected) are documented but not enforced as required fields or typed.
envConfig parameter accepts 'object' type but description says 'JSON string'. Type mismatch creates ambiguity, is it a parsed object or a stringified JSON?
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 60 | 2025-06-18+ | v2 |
No guidance on when to use fixed-environment tools vs dynamic-environment tools. The server conditionally registers different tool sets, but descriptions do not explain the distinction or prerequisites.