Skip to main content
Glama
Sign In
omniventure-agent-tools

omniventure-agent-tools

by dalnaspdal
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Hybrid

OmniVenture Edge Agent Tools (MCP & LangChain)

Sub-15ms edge intelligence tools for Claude Desktop, Cursor, LangChain, and autonomous AI agents. Reduce LLM prompt context tokens by up to 90%, repair malformed JSON in <1ms, and enforce privacy boundaries on bare-metal dual Xeon edge infrastructure.

🌐 Canonical Developer Hub & Agent Discovery: https://omniventure-api.web.app

🚀 Key Advantages for AI Agents

Tool

Capability

Token / Latency Advantage

extract_markdown

Scrapes web URLs or parses raw HTML into dense, clean Markdown without boilerplate.

Cuts context tokens by 80% to 95% (e.g. converts 60KB HTML into ~800 clean tokens). Sub-15ms.

repair_json

Fixes malformed JSON emitted by LLMs (code fences, single quotes, trailing commas, Python literals).

0.37ms edge execution. Saves 100% tokens and 2–5s latency vs re-prompting.

sanitize_pii

Redacts emails, phone numbers, payment cards, and SSNs before sending data to external public LLMs.

Sub-5ms regex masking. Enforces GDPR, HIPAA, and DLP boundaries.

distill_context

Extracts dense structured entities, financial metrics, and top keywords from documents.

Compresses 5,000+ words into ~150 structured tokens (>95% compression).

verify_email

Validates RFC 5322 syntax, verifies live DNS MX records, and flags burner/temporary domains.

Sub-15ms direct DNS MX check. Zero third-party SaaS rate limits.

lookup_ip

Resolves IP routing taxonomy, private/loopback CIDRs, and reverse PTR hostnames.

Sub-5ms native socket resolution.

Related MCP server: Fetch MCP Server

⚡ 1-Click Install: Smithery.ai

Install automatically into Claude Desktop or Cursor with the Smithery CLI:

npx -y @smithery/cli install @dalnaspdal/omniventure-agent-tools --client claude

Or for Cursor:

npx -y @smithery/cli install @dalnaspdal/omniventure-agent-tools --client cursor

⚡ Quickstart: Claude Desktop (Manual Config)

Add this to your claude_desktop_config.json:

{
  "mcpServers": {
    "omniventure-agent-tools": {
      "command": "python3",
      "args": ["-m", "distribution.mcp.fleet_runner", "--fleet", "all"],
      "env": {
        "OMNIVENTURE_GATEWAY_URL": "http://localhost:8950"
      }
    }
  }
}

⚡ Quickstart: Cursor IDE (Manual Config)

In Cursor: Settings → Features → MCP Servers → Add New MCP Server:

  • Name: omniventure-fleets

  • Type: command

  • Command: python3 -m distribution.mcp.fleet_runner --fleet all

⚡ Quickstart: LangChain & Python

from distribution.python_sdk.omniventure_tools import (
    create_markdown_tool,
    create_json_repair_tool,
    create_pii_sanitizer_tool
)

# 1. Instantiate drop-in LangChain tools
tools = [
    create_markdown_tool(api_key="YOUR_RAPIDAPI_KEY").as_langchain_tool(),
    create_json_repair_tool(api_key="YOUR_RAPIDAPI_KEY").as_langchain_tool(),
    create_pii_sanitizer_tool(api_key="YOUR_RAPIDAPI_KEY").as_langchain_tool()
]

# 2. Attach to your LangChain / CrewAI / smolagents agent
# agent = create_react_agent(llm, tools=tools)

⚡ Native RapidAPI Remote MCP Gateway

If you prefer not running local Python scripts, you can connect directly to RapidAPI's hosted MCP proxy:

npx mcp-remote https://mcp.rapidapi.com \
  --header "x-api-host: omniventure-m2m-micro-utilities.p.rapidapi.com" \
  --header "x-api-key: <YOUR_RAPIDAPI_KEY>"

💳 Settlement & Pricing

  • Endpoints are priced at £0.01 to £0.04 GBP per request.

  • Supported rails:

    • Credit Card via RapidAPI: Monthly metered billing settled to PayPal.

    • HTTP 402 / L402 Lightning: Direct programmatic preimage tokens for crypto-native agent swarms.

Marketplace Listing: RapidAPI Hub - OmniVenture M2M Micro-Utilities

Available Tools

6 tools
distill_contextA

Cognitive context distiller and entity extractor. Distills dense structured entities, numerical metrics, and salient keywords from raw text at zero cloud LLM token cost. Compresses 5,000+ words into ~150 structured tokens (>95% compression).

Inputs: raw_text (required), max_keywords.

extract_markdownA

Web-to-Markdown context compressor. Scrapes web URLs or parses raw HTML into clean, token-efficient Markdown, stripping boilerplate, scripts, ads, and navigation chrome. Cuts LLM prompt context token consumption by 80% to 95%.

Inputs: url, raw_html, include_links.

lookup_ipA

Edge IP and network taxonomy lookup. Resolves IP routing taxonomy, private/loopback/bogon CIDRs, and performs reverse DNS PTR lookups in sub-5ms latency.

Inputs: ip (required).

repair_jsonA

Universal LLM JSON repair and syntax sanitizer. Fixes malformed JSON emitted by LLMs (stripping markdown code blocks, single quotes, trailing commas, and Python literals like True/False/None) into valid RFC 8259 JSON in sub-millisecond edge time.

Inputs: raw_json (required).

sanitize_piiA

Text privacy, toxicity & PII redactor. Masks Personally Identifiable Information (emails, phone numbers, payment credit card numbers, and SSNs) from text before sending data to external public LLMs. Enforces GDPR, HIPAA, and DLP boundaries.

Inputs: text (required).

verify_emailA

B2B email and deliverability verifier. Validates RFC 5322 syntax, performs live DNS MX record verification, detects disposable/burner domains (Mailinator, TempMail), and scores corporate deliverability reputation.

Inputs: email (required), check_mx.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Deploy Server
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    TDQS

    A4.2/5.0

    Scored across 6 tools

    Disambiguation5/5

    Each tool operates on a distinct input/output pair: HTML to Markdown, malformed JSON to valid JSON, IP metadata, PII redaction, text distillation, and email verification. Although extract_markdown and distill_context both reduce text volume, their descriptions make the selection boundary clear.

    Naming Consistency5/5

    All six tool names follow the same verb_noun snake_case convention: extract, repair, lookup, sanitize, distill, and verify. The expected behavior of each tool is predictable from its name alone.

    Tool Count5/5

    Six tools is a well-scoped size for a general agent utility server. Each tool provides a discrete capability with no redundancy or obvious bloat.

    Completeness4/5

    Within each advertised utility, the tool covers the core need: extraction, repair, lookup, redaction, distillation, and verification. The set is a mixed grab bag rather than a single domain, so there is no obvious lifecycle to complete, making overall completeness slightly open-ended but not deficient.

    Resources

    Related MCP Connectors

    Related MCP Servers

    • Fetch MCP Server

      micaelmalta
      F
      license
      A
      quality
      D
      maintenance
      Reduces token consumption by 73-87% by cleaning web and API data before it reaches the LLM context window. Supports fetching URLs, searching the web, optimizing JSON, and more.
      6
      1
      -

    • token-optimizer-mcp

      ooples
      A
      license
      Not graded
      quality
      A
      maintenance
      Reduces context window usage by 60-90% through intelligent caching, compression, and smart tool replacements for Claude Code and Claude Desktop.
      4,884 npm
      578
      MIT

    • latent-gate

      KathanModh259
      F
      license
      Not graded
      quality
      B
      maintenance
      Shrinks logs, JSON, docs and web pages before your AI reads them: offline, deterministic, no API key. Folds repetitive log lines (12,037 to 127 tokens on a 400-line log) while code, URLs and quoted strings come back byte-for-byte. Also fetches URLs as clean text, counts tokens exactly, and has optional local Ollama tools for image and text compression.
      55 PyPI
      25
      -