Provides web search capabilities via the Brave Search API.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Chaprola MCP Server list my files"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP server for Chaprola — the agent-first data platform.
Gives AI agents 40 tools for structured data storage, querying, web search, URL fetching, scheduled jobs, and execution through the Model Context Protocol.
claude mcp add chaprola-mcp -e CHAPROLA_USERNAME=yourusername -e CHAPROLA_API_KEY=chp_yourkey -- npx @chaprola/mcp-serverNote: After installing, restart Claude Code to load the MCP server. The server description and tools will then be available to Claude.
Add to claude_desktop_config.json:
{
"mcpServers": {
"chaprola": {
"command": "npx",
"args": ["@chaprola/mcp-server"],
"env": {
"CHAPROLA_USERNAME": "yourusername",
"CHAPROLA_API_KEY": "chp_yourkey"
}
}
}
}Add to .vscode/mcp.json:
{
"servers": {
"chaprola": {
"command": "npx",
"args": ["@chaprola/mcp-server"],
"env": {
"CHAPROLA_USERNAME": "yourusername",
"CHAPROLA_API_KEY": "chp_yourkey"
}
}
}
}Add to .cursor/mcp.json:
{
"mcpServers": {
"chaprola": {
"command": "npx",
"args": ["@chaprola/mcp-server"],
"env": {
"CHAPROLA_USERNAME": "yourusername",
"CHAPROLA_API_KEY": "chp_yourkey"
}
}
}
}Related MCP server: Nexus-MCP
# Register (returns your API key — save it immediately)
curl -X POST https://api.chaprola.org/register \
-H "Content-Type: application/json" \
-d '{"username": "myname", "passcode": "my-secure-passcode-16chars"}'Or use the chaprola_register tool after connecting.
Tool | Description |
| Health check |
| Create account |
| Login (get new API key) |
| Check username availability |
| Delete account + all data |
| Sign Business Associate Agreement (PHI only) |
| Check BAA status |
| Get BAA text |
| Import JSON to Chaprola format |
| Get presigned upload URL |
| Process uploaded file |
| Import from URL (CSV/Excel/JSON/Parquet) |
| Export to JSON |
| List files |
| Compile .CS source to .PR bytecode |
| Execute .PR program |
| Check async job status |
| Publish program for public access |
| Remove public access |
| Run published program (no auth) |
| Run program and save output |
| Get presigned download URL |
| Filter, aggregate, join data |
| Sort data file |
| Build index on field |
| Merge two sorted files |
| HULDRA nonlinear optimization |
| Check optimization status |
| List emails |
| Read email |
| Send email |
| Delete email |
| Web search via Brave API |
| Fetch URL content as markdown/text/JSON |
| Create scheduled recurring job |
| List scheduled jobs |
| Delete scheduled job |
The server exposes reference documentation as MCP resources:
chaprola://cookbook — Language cookbook with complete examples
chaprola://endpoints — All 40 API endpoints
chaprola://auth — Authentication reference
chaprola://gotchas — Common mistakes to avoid
Variable | Required | Description |
| Yes | Your registered username |
| Yes | Your API key (format: |
Non-PHI data works without a signed BAA. If handling Protected Health Information (PHI), a human must review and sign the BAA first. The server includes guardrails that warn agents when the BAA is not signed.
Website: chaprola.org
API: api.chaprola.org
Status: UptimeRobot
Modify a data file's schema: widen/narrow/rename fields, add new fields, drop fields. Transforms existing data to match the new schema.
Inputs: name (required), project (required), add, drop, +2 more.
Check whether the authenticated user has signed the BAA
Inputs: no named parameters.
Get the current Business Associate Agreement text and version. Present to human for review before signing
Inputs: no named parameters.
Check if a username is available before registering
Inputs: username (required).
Compile Chaprola source (.CS) to bytecode (.PR). READ chaprola://cookbook BEFORE writing source. Key syntax: no PROGRAM keyword (start with commands), no commas, MOVE+PRINT 0 buffer model (not PRINT field), SEEK for primary records, OPEN/READ/WRITE/CLOSE for secondary files, LET supports one operation (no parentheses), field addressing via P.field/S.field requires primary_format/secondary_format params.
Inputs: name (required), source (required), project (required), primary_format, +1 more.
Merge a .MRG file into its parent .DA, producing a clean sorted data file. Deletes .MRG and .IGN after success. Aborts if .MRG was modified during the operation.
Inputs: file (required), project (required).
Delete an account and all associated data. Requires passcode confirmation
Inputs: passcode (required), username (required).
Delete a single record matched by a where clause. Marks the record as ignored (.IGN). Physically removed on consolidation.
Inputs: file (required), where (required), project (required).
Get a presigned S3 URL to download any file you own (1-hour expiry)
Inputs: file (required), type (required), project (required).
Delete a specific email from your mailbox
Inputs: message_id (required).
List emails in the authenticated user's mailbox
Inputs: limit, before.
Read a specific email by message_id
Inputs: message_id (required).
Send an email from your @chaprola.org address. Subject to content moderation
Inputs: to (required), text (required), subject (required), from, +1 more.
Export Chaprola .DA + .F files back to JSON
Inputs: name (required), project (required).
Run a .PR program and save output as a persistent .R file in S3
Inputs: name (required), project (required), nophi, title, +3 more.
Fetch any URL and return clean content. HTML pages converted to markdown. SSRF-protected. Rate limit: 20/day per user
Inputs: url (required), format, max_length.
Inspect a data file's schema — returns field names, positions, lengths, types, and PHI flags
Inputs: name (required), project (required).
Health check — verify the Chaprola API is running
Inputs: name.
Import JSON data into Chaprola format files (.F + .DA). Sign BAA first if handling PHI
Inputs: data (required), name (required), project (required), format, +1 more.
Import data directly from a public URL (CSV, TSV, JSON, NDJSON, Parquet, Excel). Optional AI-powered schema inference
Inputs: url (required), name (required), project (required), max_rows, +1 more.
Process a file previously uploaded to S3 via presigned URL. Generates .F + .DA files
Inputs: name (required), project (required), format.
Get a presigned S3 upload URL for large files (bypasses 6MB API Gateway limit)
Inputs: name (required), project (required).
Build an index file (.IDX) for fast lookups on a field
Inputs: file (required), field (required), project (required).
Insert a new record into a data file's merge file (.MRG). The record appears at the end of the file until consolidation.
Inputs: file (required), record (required), project (required).
List files in a project with optional wildcard pattern
Inputs: project (required), pattern.
Login and get a new API key. WARNING: invalidates the previous API key
Inputs: passcode (required), username (required).
Merge two sorted data files into one. Both must share the same format (.F)
Inputs: key (required), file_a (required), file_b (required), output (required), +1 more.
Run HULDRA nonlinear optimization using a compiled .PR as the objective evaluator
Inputs: program (required), project (required), elements (required), objectives (required), +4 more.
Check status of an async optimization job
Inputs: job_id (required), project (required).
Publish a compiled program for public access via /report
Inputs: name (required), project (required), record, primary_file.
SQL-free data query with WHERE, SELECT, aggregation, ORDER BY, JOIN, pivot, and Mercury scoring
Inputs: file (required), project (required), join, limit, +7 more.
Register a new Chaprola account. Returns an API key — save it immediately
Inputs: passcode (required), username (required).
Run a published program and return output. No auth required — program must be published first via /publish
Inputs: name (required), userid (required), project (required).
Execute a compiled .PR program. Use async:true for large datasets (>100K records)
Inputs: name (required), project (required), nophi, record, +3 more.
Run a compiled .PR program against every record in a data file. Like CHAPRPG from the original SCIOS. Use this for scoring, bulk updates, conditional logic across records.
Inputs: file (required), program (required), project (required), where, +1 more.
Check status of an async job. Returns full output when done
Inputs: job_id (required), project (required).
Create a scheduled job that runs a Chaprola endpoint on a recurring cron. Max 10 schedules/user, 15-min minimum interval
Inputs: body (required), cron (required), name (required), endpoint (required), +1 more.
Delete a scheduled job by name
Inputs: name (required).
List all scheduled jobs for the authenticated user with run history and next execution time
Inputs: no named parameters.
Search the web via Brave Search API. Returns titles, URLs, and snippets. Optional AI-grounded summary. Rate limit: 10/day per user
Inputs: query (required), count, summarize.
Sign the BAA. STOP: You MUST call chaprola_baa_text first, show the FULL text to the human, and get their EXPLICIT typed approval before calling this. Never sign automatically. Only needed for PHI — non-PHI data works without a BAA.
Inputs: signatory_name (required), organization, signatory_title.
Sort a data file by one or more fields. Modifies the file in place
Inputs: file (required), project (required), sort_by (required).
Remove public access from a published program
Inputs: name (required), project (required).
Update fields in a single record matched by a where clause. If no sort-key changes, updates in place; otherwise marks old record ignored and appends to merge file.
Inputs: set (required), file (required), where (required), project (required).
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 44 tools
Several tools have overlapping purposes, particularly the import variants (import, import_process, import_download, import_url) which all involve bringing data in but via different methods. Also, chaprola_report and chaprola_export_report both run programs, and chaprola_run and chaprola_run_each are similar. However, most tools have clear distinct roles (e.g., insert_record vs update_record vs delete_record), and descriptions mitigate some ambiguity.
All tools share the consistent 'chaprola_' prefix, which is good. However, the suffix pattern is mixed: some tools use verb_noun (insert_record, schedule_delete), others use bare verbs (import, export, list), and there are inconsistent variations like import_process vs import_download vs import_url. Also, BAA tools place the domain word differently (baa_text, sign_baa, baa_status). This inconsistency reduces predictability.
With 44 tools, this server is significantly over the typical well-scoped range of 3-15 tools. The large count likely overwhelms agents and makes tool selection harder. Many tools could be consolidated (e.g., the import variants) or omitted (e.g., email and web search may be tangential). While the platform is broad, the sheer number makes the server unwieldy.
The tool surface is quite comprehensive for the Chaprola data platform, covering import/export, record CRUD, schema operations, program compilation/execution, publishing, scheduling, email, and web search. Minor gaps include a direct 'get record' operation (though query covers it) and file deletion tools. Overall, the domain is well-covered with only a few workarounds needed.
Persistent memory and knowledge management for AI agents with semantic search and 50+ tools.
Context and control for any AI. Agent identity, scoped access, and proof across tools.
Direct access to 60+ scraping and search tools. Extract structured data from Google (Search, Maps, Trends), Amazon, Airbnb, Social Media, and any web page directly into your AI agent.
Persistent memory for AI agents: save, search and connect context across tools and sessions.
