quap-mcp tools
Tools quap-mcp exposes to MCP clients such as Cursor and Claude Desktop.
These are the tools an MCP client sees when quap-mcp is on. The list is read from quap-mcp itself each time the documentation is built, so it always matches the shipped version.
Highlights:
run_planandlist_plan_opsexpose every operation of Quap's AI assistant — filter, mask, sort, reshape, analyse, convert, generate SQL and code — as validated steps your client's model can call.analyzeandprofile_documentlet a model understand a dataset (record arrays, field kinds) without reading every value.sql_transpile,sql_to_linqandlinq_to_sqltranslate queries between dialects and C# LINQ.
All tools
| format_json | Pretty-print a JSON string via Quap's format pipeline. |
| minify_json | Minify a JSON string to a single line. |
| validate_json | Validate whether a string is valid JSON. |
| convert_format | Convert between json, ndjson, yaml, and xml (pretty or compact). |
| list_transforms | List available text transforms (case, codec, hash, crypto, …) with metadata. |
| apply_transform | Apply a Quap text transform by id (same registry as the desktop Transform menu). |
| sql_detect | Detect the likely SQL product dialect for a SQL or LINQ snippet. |
| sql_transpile | Transpile SQL between product dialects (or SQL ↔ LINQ). |
| format_sql | Pretty-print SQL (uppercase keywords, 2-space indent). |
| minify_sql | Collapse SQL to a compact single-line form. |
| sql_preview | Parse SQL into a preview model (statements, joins, CREATE/INSERT structures). |
| generate_code | Generate C#/Java/TypeScript/Python types from JSON or XML. |
| dedupe | Deduplicate a JSON array or plain-text lines (preserve order). |
| diff | Line-based diff of two strings (unified-style markers). |
| line_set | Line set ops: intersection, union, diffAB, diffBA, or duplicates report. |
| structural_json_diff | Structural JSON diff (add/remove/change by path). |
| openapi_inspect | Detect/validate OpenAPI and return suggestions + endpoints markdown. |
| parse_curl | Parse a curl command into method, URL, headers, and body. |
| explain_regex | Explain a regular expression as structured parts (and validity). |
| build_regex | Build a simple regex from a high-level spec (digits/letters/email/url/literal). |
| list_plan_ops | List every operation run_plan can perform for a document format (the same functions Quap's AI assistant uses: mask, filter, sort, dedupe, delete, set_value, randomize, transform, format_value, replace, convert, analyze, sql_script, codegen, sql_transpile, sql_to_linq, openapi, stats…), with the JSON schema of the steps. |
| run_plan | Run one or more Quap operations on a document locally, in order (each step works on the previous step's result). Steps are validated first. Returns the output text, its format, per-step status/matches/messages and, for a final analyze step, the result table. Use list_plan_ops for the available steps. |
| analyze | Group / aggregate / filter an array of records locally (count, sum, avg, min, max, median, percentiles; date and range buckets) and get a table plus a chart suggestion. |
| profile_document | Describe a document's record arrays and fields (kind: number, date, category, email, phone, tags…; empty / distinct counts, sensitive key names) without returning values. |
| generate_sql_script | SQL script from data: CREATE TABLE statements (nested objects / arrays become related tables with foreign keys), optionally with INSERTs. |
| sql_to_linq | Translate an SQL query to C# LINQ (Entity Framework). |
| linq_to_sql | Translate C# LINQ query syntax to SQL. |
| document_stats | Count nodes, keys (and distinct keys), objects, arrays, value types and empty values of a structured document. |
| search_document | Find lines in content that contain a query substring. |