Helve
Pay-per-call AI tool gateway (313 tools, 255 providers) with spend_limit_usd caps & $0 failed calls
Should I use this
Quality & Safety
Based on automated analysis of tool definitions and protocol compliance.
Context Cost
This is the approximate number of tokens consumed each time the server's tools are loaded into a model's context. Higher counts reduce the attention available for other tasks.
Install
One-Click Install
Add this to your `claude_desktop_config.json` file:
{
"mcpServers": {
"helve": {
"url": "https://helve.dev/mcp"
}
}
}Remote endpoints
https://helve.dev/mcpstreamable-httpWhat it can do
Tool inventory
Tools (11)
🟢search_tools(query, category, limit)
Search or browse Helve's broad catalog of third-party tools (LLMs, embeddings, web search, extraction, site mapping, deep research, TTS/audio, SEC filings, GitHub, arXiv, academic papers, weather, geocoding, DNS, FX, package registries). Omit `query` (or pass `category` alone) to browse live tools without a keyword. Returns tool IDs, JSON input schemas, examples, and pricing hints for immediate invocation via call_tool.
Input Schema
{
"type": "object",
"properties": {
"query": {
"description": "Optional natural-language capability or tool you need (omit to browse live tools)",
"type": "string"
},
"category": {
"description": "Optional category filter (search, extract, crawl_map, research_answer, llm, embeddings, audio_tts, transcription, finance, developer_data, knowledge_geo, utilities)",
"type": "string"
},
"limit": {
"description": "Maximum tools to return (default 8, or 12 when browsing)",
"type": "integer",
"minimum": 1,
"maximum": 30
}
},
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢list_categories
Browse Helve's live tool catalog by category without guessing a search keyword. Returns every category, its live tool count, and sample tool IDs.
Input Schema
{
"type": "object",
"properties": {},
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢get_balance
Check your Helve account USD credit balance, 30-day call & spend usage summary, and the active API key's spend_limit_usd budget cap.
Input Schema
{
"type": "object",
"properties": {},
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢call_tool(tool_id, arguments)
Invoke any tool in Helve's catalog by tool_id with tool-specific arguments (discover tool_ids and schemas first via search_tools). Metered per call against your single Helve balance; failed upstream calls debit $0.
Input Schema
{
"type": "object",
"properties": {
"tool_id": {
"type": "string",
"minLength": 1,
"description": "Catalog tool id from search_tools, e.g. 'openrouter.chat', 'perplexity.answer', 'sec.edgar_search', 'elevenlabs.tts'"
},
"arguments": {
"default": {},
"description": "Tool arguments matching the tool's input_schema",
"type": "object",
"propertyNames": {
"type": "string"
},
"additionalProperties": {}
}
},
"required": [
"tool_id"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}⚪request_tool(tool_name, use_case, provider, category, docs_url)
Log a third-party tool or API you need that is not yet in Helve's catalog. Helve tracks agent demand and prioritizes integrations where agents are blocked.
Input Schema
{
"type": "object",
"properties": {
"tool_name": {
"type": "string",
"minLength": 1,
"maxLength": 200,
"description": "Name of the API, SaaS, or tool you need (e.g. 'Resend', 'Browserbase', 'Apollo.io')"
},
"use_case": {
"type": "string",
"minLength": 1,
"maxLength": 1000,
"description": "What you are trying to accomplish with it"
},
"provider": {
"type": "string",
"maxLength": 120
},
"category": {
"type": "string",
"maxLength": 100
},
"docs_url": {
"type": "string",
"maxLength": 500,
"format": "uri"
}
},
"required": [
"tool_name",
"use_case"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢search(query, provider, max_results, freshness, include_domains, ...)
Search the live web. One call searches across Helve's providers (Exa, Parallel, Tavily, Brave, Serper, Firecrawl, Linkup, Valyu, Jina) and returns titles, URLs, publication dates, and content snippets. provider "fusion" queries several engines in parallel and ranks by cross-engine consensus — use it when result quality matters most. Use freshness for recent events and include_domains to stay on specific sites.
Input Schema
{
"type": "object",
"properties": {
"query": {
"type": "string",
"minLength": 1,
"description": "What to search the web for, in natural language"
},
"provider": {
"description": "Search engine to use; omit for automatic routing",
"type": "string",
"enum": [
"auto",
"fusion",
"exa",
"parallel",
"tavily",
"brave",
"serper",
"firecrawl",
"linkup",
"valyu",
"jina",
"octen",
"perplexity"
]
},
"max_results": {
"description": "How many results (default 10)",
"type": "integer",
"minimum": 1,
"maximum": 20
},
"freshness": {
"description": "Only results published within this window",
"type": "string",
"enum": [
"day",
"week",
"month",
"year"
]
},
"include_domains": {
"description": "Restrict results to these domains",
"maxItems": 10,
"type": "array",
"items": {
"type": "string"
}
},
"category": {
"type": "string",
"enum": [
"general",
"news",
"company",
"people",
"research"
]
}
},
"required": [
"query"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟡transcribe(audio_url, provider, language, diarize, keyterms, ...)
Submit a prerecorded audio or video URL for durable batch transcription. Returns immediately with a job ID; use transcription_status to retrieve its state and completed result.
Input Schema
{
"type": "object",
"properties": {
"audio_url": {
"type": "string",
"format": "uri",
"description": "HTTPS URL of the recording, fetched by the provider. Signed URLs are fine; it must stay reachable until the provider has read it. No embedded credentials or fragments."
},
"provider": {
"description": "Which speech-to-text provider runs the job. `auto` (default) picks the first configured provider. The choice is persisted on the job before execution.",
"type": "string",
"enum": [
"assemblyai",
"deepgram",
"elevenlabs",
"auto"
]
},
"language": {
"description": "BCP-47-style language tag such as `en` or `en-US`. Omit to let the provider detect the language.",
"type": "string",
"pattern": "^[a-zA-Z]{2,3}(?:-[a-zA-Z0-9]{2,8})*$"
},
"diarize": {
"description": "Label speakers in `words` as `speaker_0`, `speaker_1`, … in order of first appearance. Default false.",
"type": "boolean"
},
"keyterms": {
"description": "Recognition hints: names, jargon, product terms. Supported on AssemblyAI, Deepgram Nova-3, and ElevenLabs Scribe v2; other models ignore them with a warning.",
"minItems": 1,
"type": "array",
"items": {
"type": "string"
}
},
"timestamps": {
"description": "`word` (default) returns per-word timings in `result.words`; `none` omits them.",
"type": "string",
"enum": [
"word",
"none"
]
},
"strict": {
"description": "When true, any option the selected provider or model cannot honour rejects the request instead of producing a warning. Default false.",
"type": "boolean"
},
"provider_options": {
"description": "Provider-native options, validated by the adapter. Model selection lives here. Transport settings such as URLs, callbacks, and authentication cannot be overridden.",
"type": "object",
"properties": {
"assemblyai": {
"description": "AssemblyAI-native options, for example `{ \"model\": \"best\" }`.",
"type": "object",
"propertyNames": {
"type": "string"
},
"additionalProperties": {}
},
"deepgram": {
"description": "Deepgram-native options, for example `{ \"model\": \"nova-3\" }`.",
"type": "object",
"propertyNames": {
"type": "string"
},
"additionalProperties": {}
},
"elevenlabs": {
"description": "ElevenLabs-native options, for example `{ \"model_id\": \"scribe_v2\" }`.",
"type": "object",
"propertyNames": {
"type": "string"
},
"additionalProperties": {}
}
},
"additionalProperties": false
}
},
"required": [
"audio_url"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢transcription_status(job_id)
Get the current state of a durable transcription job. Includes the normalized transcript when complete.
Input Schema
{
"type": "object",
"properties": {
"job_id": {
"type": "string",
"pattern": "^job_[0-9a-f-]{36}$"
}
},
"required": [
"job_id"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢list_transcriptions(limit)
List your recent durable audio transcription jobs and their statuses.
Input Schema
{
"type": "object",
"properties": {
"limit": {
"description": "Maximum jobs to return (default 20)",
"type": "integer",
"minimum": 1,
"maximum": 50
}
},
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🔴cancel_transcription(job_id)
Cancel a queued or running durable transcription job owned by your account.
Input Schema
{
"type": "object",
"properties": {
"job_id": {
"type": "string",
"pattern": "^job_[0-9a-f-]{36}$"
}
},
"required": [
"job_id"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢extract(urls, query, provider, max_chars)
Extract clean, readable content from one or more web page URLs (articles, docs, listings). Pass query to pull only the sections relevant to a topic instead of the whole page. Content is markdown.
Input Schema
{
"type": "object",
"properties": {
"urls": {
"minItems": 1,
"maxItems": 10,
"type": "array",
"items": {
"type": "string",
"format": "uri"
},
"description": "The pages to extract"
},
"query": {
"description": "Optional focus: extract only content relevant to this",
"type": "string"
},
"provider": {
"type": "string",
"enum": [
"auto",
"exa",
"tavily",
"parallel",
"serper",
"firecrawl",
"linkup",
"valyu",
"jina",
"octen"
]
},
"max_chars": {
"description": "Max characters of content per URL (default 8000)",
"type": "integer",
"minimum": 500,
"maximum": 50000
}
},
"required": [
"urls"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}Community
Evidence