Skip to main content
Glama

Exa

by io.github.pipeworx-io

Search

search
Read-onlyIdempotent

Neural/semantic web search — find pages by meaning, not just keywords. Returns title, URL, published date, author, and relevance score. Optionally retrieve clean page text inline. Example: search({ query: "startups building AI agents for customer support", num_results: 10, type: "neural" })

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
typeNoSearch mode: 'neural' (semantic/meaning-based), 'keyword' (traditional), or 'auto' (Exa picks). Default 'auto'.
queryYesThe search query — describe what you want by meaning, e.g. "recent research on protein folding with diffusion models"
_apiKeyNoOptional — your own Exa API key for higher limits; omit to use the shared Pipeworx key.
categoryNoOptional focus category, e.g. 'company', 'research paper', 'news', 'pdf', 'github', 'tweet'.
num_resultsNoNumber of results to return (default 10, max 25)
include_textNoIf true, include the clean parsed page text (up to 2000 chars) for each result. Default false.

Schema Changelog

Changes observed during successful MCP inspections. Dates show when Glama detected each change.

  1. Changed1 schema field changed
    • addedInput schema / examples
      Added value: +[
+  {
+    "num_results": 10,
+    "query": "recent research on protein folding with diffusion models",
+    "type": "neural"
+  },
+  {
+    "category": "company",
+    "include_text": true,
+    "num_results": 5,
+    "query": "startups building AI agents for customer support",
+    "type": "neural"
+  }
+]
  2. First observed

TDQS

A4.3/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint, openWorldHint, idempotentHint, and destructiveHint. The description adds value by specifying return fields (title, URL, date, author, score) and the option to include clean page text, providing context beyond the annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is concise (3 sentences plus an example) and front-loaded with the core purpose. Every sentence adds value, with no wasted words.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the complexity (6 parameters, no output schema), the description adequately covers return fields and optional text retrieval. It is sufficient for an agent to correctly invoke the tool.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the schema already documents all parameters. The description includes an example that reinforces usage but does not add new semantic information beyond what the schema provides.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states it performs neural/semantic web search, distinguishes from keyword search, and lists return fields. It differentiates from sibling tools like search_within by emphasizing meaning-based search.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides a clear example and states the tool is for finding pages by meaning. It implicitly distinguishes from keyword search via the 'type' parameter, but does not explicitly state when not to use or list alternatives.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Exa toolsInspect
Try in Browser
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

A3.9/5.0
Disambiguation3/5

Several tools share the same basic purpose: ask_pipeworx, ask_pipeworx_beta, ask_pipeworx_grounded, and deep_research all route questions to data sources, and ask_pipeworx_beta is currently identical to ask_pipeworx. The polymarket_* family has five overlapping tools (edges, arbitrage, edge_tracker, fill_risk, kalshi_spread), though detailed descriptions and explicit 'use when' guidance help separate them. Overall, an agent can generally pick the right tool but faces real ambiguity in the query-router and betting clusters.

Naming Consistency4/5

Most tools follow a clear verb_noun snake_case convention (search, get_contents, resolve_entity, validate_claim, subscribe, unsubscribe). However, several noun-first names break the pattern: entity_profile, ai_visibility_check, pipeworx_feedback, pipeworx_trending, and the polymarket_* family, plus adjective-noun names like recent_alerts and recent_changes. The deviations are readable and mostly clustered around product-specific domains, so the inconsistency is minor.

Tool Count2/5

At 34 tools, this significantly exceeds the 25+ threshold for a heavy tool surface. The server bundles four distinct domains — web search, structured data routing, prediction-market analysis, and memory/subscriptions — into one MCP endpoint, which inflates the count. While each domain has some justification, a more focused split into separate servers would yield better coherence.

Completeness4/5

The surface is remarkably thorough for its blended scope: search has query/retrieve/similar/within, structured data has default/grounded/beta/deep-research modes, subscriptions have full lifecycle coverage, and memory has save/recall/delete. Minor gaps exist, such as no subscription-update tool and no direct pipeworx:// URI reader in the tool list, but these are workable. The prediction-market and entity-analysis workflows are covered end to end.

Resources