Search and research MCP for knowledge work.

One server to search the web and specialised indexes built for knowledge work, and run multi-step research with cited outputs, from Claude, Codex or any agent. Read the docs.

Works withClaudeCodexCursorHermes Agent (Nous Research)Hermespiand more…

Connect

Claude

Settings → Connectors → Add custom connector. Paste the endpoint and sign in with Valyu.

endpoint
https://mcp.valyu.ai/mcp

ChatGPT

Settings → Connectors → Create. Paste the endpoint and sign in with Valyu.

endpoint
https://mcp.valyu.ai/mcp

Claude Code

terminal
claude mcp add --transport http valyu https://mcp.valyu.ai/mcp

Codex, Cursor and others

Any client that speaks Streamable HTTP. Sign in with Valyu when prompted, or put an API key on the URL (an Authorization: Bearer header works too):

mcp.json
{
  "mcpServers": {
    "valyu": {
      "type": "http",
      "url": "https://mcp.valyu.ai/mcp?valyuApiKey=<VALYU_API_KEY>"
    }
  }
}

Hermes Agent (Nous Research)Hermes, pi and terminal agents

Point the agent at the endpoint; it signs in on first use. Keys live at platform.valyu.ai.

Tools

valyu_searchSearch the web and specialised indexes built for knowledge work in one call, or scope to one index.

Any question that needs current information, primary sources or citations. Returns ranked results with content, URLs and relevance; scope with included_sources for precision, leave it unscoped for breadth.

Parameters
queryrequiredstring

Natural language query.

included_sourcesstring[]

ADVANCED, and a hard filter: everything not named is excluded, so a wrong value returns less than no filter at all.

source_biasesobject

ADVANCED, but the safe one: it only reorders, so it cannot empty the result set.

excluded_sourcesstring[]

ADVANCED.

max_num_resultsinteger = 10

How many distinct documents to return.

response_lengthinteger = 4000

Maximum characters returned per result (500-100000).

start_datestring

ADVANCED, and the option most likely to return nothing by accident.

end_datestring

ADVANCED.

include_imagesboolean = false

Return figures extracted from each result, as links.

country_codestring

ADVANCED.

valyu_datasourcesFind datasets by describing what you need; returns the ids that scope a search.

Call before a scoped search to see which specialist datasets exist for a topic. Returns dataset ids, what each covers, and whether the plan includes it.

Parameters
querystring

Describe the data you need in plain language, e.g. "peer-reviewed medical literature", "FDA drug labels", "insider transactions", "CPI inflation".

check_accessboolean = true

Annotate each dataset with whether the current plan includes it.

categorystring

v1 option: only list datasets in this category id.

valyu_contentsRead any URL as clean markdown.

Read one or more pages in full. Returns clean markdown, optionally summarised to an instruction.

Parameters
urlsrequiredstring[]

1-10 absolute http(s) URLs to fetch.

summaryboolean | string

true for a general summary, or a string instruction describing what to extract, e.g. "only the methodology and results".

extract_effort"normal" | "high" | "auto" = "auto"

"normal" is fastest; "high" renders JavaScript and is slower but succeeds on SPAs and difficult pages; "auto" picks per URL.

screenshotboolean = false

Also return a full-page screenshot of each URL, as a link.

response_lengthinteger = 30000

Maximum characters returned per page (500-200000).

valyu_deepresearchStart a long-running deepresearch task that writes a cited report.

Only when the user wants a written deliverable: a report, literature review, due diligence. Runs for minutes to hours as a research agent; returns a task id immediately.

Parameters
querystring

The deepresearch brief.

mode"fast" | "standard" | "heavy" | "max" = "fast"

Depth/cost tier.

wait_secondsinteger = 0

Block for up to this many seconds waiting for completion before returning a task id.

report_formatstring

Natural-language guidance for the report structure, e.g. "a one-page memo with a risks table" or "bullet points, no preamble".

research_strategystring

Natural-language guidance for how to run the research, e.g. "prioritise primary sources and SEC filings over news".

included_sourcesstring[]

ADVANCED, and a hard filter.

excluded_sourcesstring[]

ADVANCED.

source_biasesobject

ADVANCED, but the safe one: it only reorders and cannot starve the report.

search_type"all" | "web" | "proprietary"

ADVANCED.

country_codestring

ADVANCED.

start_datestring

ADVANCED, and the easiest way to starve a report by accident.

end_datestring

ADVANCED.

urlsstring[]

Up to 10 URLs the agent must read and use as context (e.g. a filing, a trial page, a competitor's site).

previous_reportsstring[]

Up to 3 earlier task ids whose reports become context for this one - for follow-ups and updates.

filesobject[]

Up to 10 files (as base64 data URLs) for the agent to analyse.

pdfboolean = false

Also produce a PDF of the report (a download link is returned with the report).

output_schemaobject

JSON Schema for structured JSON output instead of a markdown report.

deliverablesstring | object[]

Up to 10 files to generate from the report, e.g. "an Excel sheet of every trial with phase and sponsor" or {"type": "pptx", "description": "10-slide summary"}.

toolsobject

Enable or disable optional agent tools.

hitlobject

Human-in-the-loop checkpoints.

mcp_serversobject[]

Up to 5 external MCP servers whose tools the deepresearch agent may call (e.g. an internal data source).

webhook_urlstring

HTTPS URL to POST to when the task completes (HMAC-SHA256 signed).

alert_emailstring

Email the finished report link to this address (must belong to the user's organisation).

metadataobject

Arbitrary key/value tags stored with the task.

workflow_idstring

Beta: run a saved Valyu workflow (prompt template) instead of a free-form query.

workflow_paramsobject

Values for the workflow's template variables, keyed by variable key.

workflow_versioninteger

Pin a specific workflow version; defaults to the current one.

valyu_deepresearch_statusCheck a task, collect the finished report, or list recent tasks.

Check a running task, collect the finished report with citations, or, with no task id, list recent tasks.

Parameters
task_idstring

The task id returned by valyu_deepresearch.

limitinteger = 20

Listing only: how many recent tasks to return (default 20).

wait_secondsinteger = 0

0 (default) returns the current status immediately - right for a conversation: report it and let the user come back later.

include_sourcesboolean = true

Include the citation list with deep links into each source (completed tasks only).

valyu_deepresearch_controlSteer, answer a checkpoint, share, cancel or delete a task.

Redirect a running task, answer a checkpoint it is waiting on, toggle the share link, or cancel or delete it.

Parameters
task_idrequiredstring

The task id returned by valyu_deepresearch (or listed by valyu_deepresearch_status).

actionrequired"steer" | "cancel" | "respond" | "share" | "delete"

"steer", "cancel", "respond", "share" or "delete" - see the tool description.

instructionstring

The steering instruction, in plain language.

interaction_idstring

The checkpoint id shown by valyu_deepresearch_status.

responseobject

The checkpoint answer, in the shape valyu_deepresearch_status specified for that checkpoint type.

publicboolean

Required when action is "share": true publishes a share link, false removes it.

Sources

The live web plus academic, financial, clinical, patent, economic, legal and security datasets, growing every month. Browse every data source.