RquestSAFE
A MCP server providing realistic browser-like HTTP request capabilities with accurate TLS/JA3/JA4 fingerprints for bypassing anti-bot measures. It also supports converting PDF and HTML documents to Markdown for easier processing by LLMs.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://pypi.org/project/mcp-rquest/) [](https://pypi.org/project/mcp-rquest/) [](https://github.com/xxxbrian/mcp-rquest) [](https://github.com/xxxbrian/mcp-rquest)
A Model Context Protocol (MCP) server that provides advanced HTTP request capabilities for Claude and other LLMs. Built on rquest, this server enables realistic browser emulation with accurate TLS/JA3/JA4 fingerprints, allowing models to interact with websites more naturally and bypass common anti-bot measures. It also supports converting PDF and HTML documents to Markdown for easier processing by LLMs.
Features
- Complete HTTP Methods: Support for GET, POST, PUT, DELETE, PATCH, HEAD, OPTIONS, and TRACE
- Browser Fingerprinting: Accurate TLS, JA3/JA4, and HTTP/2 browser fingerprints
- Content Handling:
- Automatic handling of large responses with token counting
- HTML to Markdown conversion for better LLM processing
- PDF to Markdown conversion using the Marker library
- Secure storage of responses in system temporary directories
- Authentication Support: Basic, Bearer, and custom authentication methods
- Request Customization:
- Headers, cookies, redirects
- Form data, JSON payloads, multipart/form-data
- Query parameters
- SSL Security: Uses BoringSSL for secure connections with realistic browser fingerprints
Available Tools
- HTTP Request Tools:
http_get- Perform GET requests with optional parametershttp_post- Submit data via POST requestshttp_put- Update resources with
ab3c109d8827OBSERVED · 2026-10-08Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.
claude mcp add mcp-rquest -- uvx mcp-rquest
{
"mcpServers": {
"mcp-rquest": {
"command": "uvx",
"args": [
"mcp-rquest"
]
}
}
}Exposed tools (12)
7 read · 4 write · 1 destructive. Blast radius: 1 tool can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
get_model_state | read | Get the current state of the PDF models(used by `get_stored_response_with_markdown`) loading process |
get_stored_response | read | Retrieve a stored HTTP response by its ID |
get_stored_response_with_markdown | read | Retrieve a stored HTTP response by its ID and convert it to Markdown format. Supports HTML and PDF content types. (Converting large PDF to Markdown may cause timeout, just wait and try again.) |
http_delete | destructive | Make an HTTP DELETE request to the specified URL |
http_get | read | Make an HTTP GET request to the specified URL |
http_head | read | Make an HTTP HEAD request to retrieve only headers from the specified URL |
http_options | read | Make an HTTP OPTIONS request to retrieve options for the specified URL |
http_patch | write | Make an HTTP PATCH request to the specified URL |
http_post | write | Make an HTTP POST request to the specified URL |
http_put | write | Make an HTTP PUT request to the specified URL |
http_trace | read | Make an HTTP TRACE request for diagnostic tracing of the specified URL |
restart_model_loading | write | Restart the PDF models(used by `get_stored_response_with_markdown`) loading process if it failed or got stuck |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (2 observation(s))
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (3)
http_delete
.bumpversion.toml
pdf_content = base64.b64decode(content)
Gates applied: no_behavioural_pass.
ab3c109d8827full audit observations/trust-audit/mcp-server/xxxbrian__rquest.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | ab3c109d8827 | SAFE | B | 89 | first audit |
Questions
What is the Rquest MCP server?
A MCP server providing realistic browser-like HTTP request capabilities with accurate TLS/JA3/JA4 fingerprints for bypassing anti-bot measures. It also supports converting PDF and HTML documents to Markdown for easier processing by LLMs.
What tools does Rquest expose?
12 in total: 7 read-only, 4 that write, and 1 that can delete or overwrite (http_delete). Every one is listed on this page with its risk.
Is Rquest safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B. Separately from the audit: 1 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does Rquest need?
No credential environment variables were found in its source, so it appears to need none.
How does Rquest run?
It speaks stdio, so it runs as a local process your client starts. It is published on PyPI as mcp-rquest.
How current is this page?
The grade is for one exact copy of the source (ab3c109d8827), read on 2026-10-08. The repository is watched and re-audited when it changes.