For Agents
Enrich rows of company or contact data with AI web research, build datasets from a prompt, scrape pages, run quick web searches, and poll or stop the resulting runs.
Use for: Find the tech stack for each company in this list, Build a list of the top 50 US SaaS companies with their websites, Scrape this pricing page and return the text, Search the web for news about a company from the last month
Not supported: Does not send outreach emails, store CRM records, or verify email deliverability. Use for AI web research, enrichment, dataset building, and scraping only.
The Riveter API fills in new columns for rows of input data using AI, web searches, and web scrapes, so a list of companies can come back with their tech stack, pricing structure, or security tools. It also builds datasets of companies, people, or products from a natural-language prompt or a structured spec, scrapes a URL to plain text, runs a synchronous web search with structured results, and answers single research questions through its search agent. Asynchronous work returns a run that is checked, fetched, and stopped through one set of run endpoints, with optional webhook delivery when it finishes. Monitors re-run a saved enrichment daily, weekly, or monthly, and extractions turn a website into a reusable structured scraping recipe.
Install Jentic One Beta
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the Riveter API, or any other public or private API you need. You set the rules, the agent never sees your credentials, and every call is logged.
Two steps, two machines. Install the instance in a safe environment, then register your agent from wherever it runs.
Step 1: Jentic One Host machine
# On the machine that will host your Jentic One instance:
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Friveterhq.com%2Friveterhq" | shStep 2: Agent machine
# On the machine where your agent runs (keep this separate from the instance):
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Friveterhq.com%2Friveterhq" | sh
jentic register # connects your agent to your Jentic One instanceJentic One is in public beta. The setup above keeps your agent separate from the instance, which is what you want before using real credentials: an agent running as the same OS user as Jentic One can read its stored keys directly. Just evaluating? A single local install is fine to start. See the secure deployment guide for the tiers.
What an agent can do with Riveter API.
Enrich rows of input data with new AI-researched columns through /enrich, using a saved enrichment_id or an inline prompt and attributes
Build a dataset of companies, people, or products from a natural-language prompt, identifiers, and qualifiers
Scrape a webpage to text synchronously, with an optional proxy_country_code
Run a quick web search with an optional date_start and date_end and get structured results in one request
Ask the search agent a single research question and get an answer shaped by an optional output_schema
Check, fetch, and stop runs, with the wait parameter long-polling for results
Schedule a monitor that re-runs an enrichment daily, weekly, or monthly and posts results to a webhook
Patterns agents use Riveter API for, with concrete tasks.
★ AI research agent that enriches account lists
An AI agent preparing sales research can hand a list of company names to Riveter instead of browsing each website itself. Through Jentic the agent searches for 'enrich a list of companies', loads the Riveter enrich operation with its input schema, and posts the rows with a prompt and the attributes it wants back, with the bearer key added at call time. It then long-polls the run result and returns the filled-in table.
POST /enrich with a list of 20 company domains in input and attributes for tech stack and pricing model, then call GET /runs/{id}/result with wait=30 until output is returned
Prompt-built prospect datasets
Instead of buying a static list, a team describes the rows it wants, such as US fintech companies with a public API, and Riveter's dataset builder finds them. The request can combine a prompt with identifiers, qualifiers, attributes, and a max_items cap, and auto_run_enrichment can chain an enrichment over the new rows. A dataset can later be extended with more rows.
POST /datasets with the prompt 'fintech companies in the US with a public developer API' and max_items 50, then fetch the finished run with GET /runs/{id}/result
Scheduled change monitoring
Some attributes change over time, such as a competitor's pricing or headcount. A Riveter monitor re-runs a saved enrichment over fixed input rows on a daily, weekly, or monthly cadence at a set hour and timezone, and can post results or only changes to a webhook. Listing a monitor's runs shows each scheduled execution and its status.
POST /monitors with an enrichment_id, cadence weekly, hour 9, minute 0, timezone Europe/Dublin, and a webhook_url for change alerts
Quick search and page scraping for grounding
An agent answering a factual question often needs one search and one page read, not a full enrichment. Riveter's quick search returns structured results with URLs, titles, and snippets in the same response, and the scrape endpoint returns a page as text synchronously. The search agent covers scoped one-off questions such as a company's NAICS code.
POST /quick_search with a query about a company's latest funding round, then POST /scrape on the top result URL and summarise the page
27 endpoints — the riveter api fills in new columns for rows of input data using ai, web searches, and web scrapes, so a list of companies can come back with their tech stack, pricing structure, or security tools.
METHOD
PATH
DESCRIPTION
/enrich
Enrich rows of input data
/runs/{id}/result
Fetch a run's output
/runs/{id}
Check run status
/datasets
Build a dataset
/quick_search
Run a quick web search
/scrape
Scrape a webpage
/search_agent
Ask a research question
/monitors
Create a monitor
/enrich
Enrich rows of input data
/runs/{id}/result
Fetch a run's output
/runs/{id}
Check run status
/datasets
Build a dataset
/quick_search
Run a quick web search
/scrape
Scrape a webpage
/search_agent
Ask a research question
/monitors
Create a monitor
This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use.
Base layer of spec validity and structural soundness.
Aggregated quality score from linter diagnostics, weighted by severity.
Percentage of `$ref` references that resolve successfully.
Checks whether the API description parses successfully and conforms to its declared specification (e.g., OpenAPI).
Structural correctness score based on schema issues using logarithmic dampening.
Clarity, completeness, and ingestion readiness for developers and tooling.
How richly the API is illustrated with examples.
Percentage of examples that conform to their schemas.
Percentage of operations with complete response definitions (success, client error, server error).
Health of API ingestion, bundling, and resolution within Jentic pipelines.
Semantic breadth, depth, and agent comprehension for AI systems.
Coverage of descriptions across API elements.
Coverage of RFC 9457 Problem Details for error responses.
Coverage, uniqueness, and casing consistency of operationIds for AI inference.
Coverage of summaries across operations/tags/info.
Functional utility, complexity comfort, and AI orchestration readiness.
Agent comfort level based on API operational and structural complexity.
Trust, risk posture, and security compliance.
Average quality of security schemes based on authentication method strength (weakest link for OAuth2).
Findability, semantic richness, and reasoning readiness.
Clarity and depth of descriptions across API elements.
Score it yourself
Every API in the directory is allowlisted, so you can re-score it with no key required.
npx @jentic/api-scorecard-cli score <openapi-url>What agents get from Jentic-routed access to this vendor.
Setup
Wiring Riveter by hand means sending a bearer key, choosing between inline configs and saved enrichment ids, and handling the run lifecycle with long-polling or webhooks. Through Jentic you install once, import Riveter from the API Directory, store the key once, and your agent calls it.
Permission scoping
Riveter carries the rows, enrichment_id, and prompt inside the request body of calls like /enrich, so rules bound which operations your agent may call, not which enrichment it runs. You choose the operations it may call, so credit-spending ones like /enrich, /datasets, or creating monitors are not included unless you add them.
Credential isolation
Your Riveter API key is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context.
Intent-based discovery
Agents search Jentic by intent such as 'enrich a list of companies' or 'scrape a webpage', and Jentic returns the matching Riveter operation with its input schema so the agent calls the right endpoint without browsing the reference docs.
Alternatives and complements available in the Jentic catalogue.
Specific to using Riveter API through Jentic.
What authentication does the Riveter API use?
Per its OpenAPI spec, the Riveter API uses HTTP bearer authentication: send your API key as `Authorization: Bearer YOUR_API_KEY` on each request. Through Jentic the key is held by your own instance and injected at call time, so the agent never sees it.
Can I get enrichment results without polling with the Riveter API?
Yes. Pass webhook_url in the body when you start a run with /enrich and Riveter posts the results to your URL when the run finishes; dataset builds take dataset_webhook_url. If you poll instead, GET /runs/{id}/result accepts a wait parameter that long-polls for the output.
What are the rate limits for the Riveter API?
The Riveter OpenAPI spec states a default of 30 requests per minute per endpoint group, with `X-RateLimit-Limit`, `X-RateLimit-Remaining`, and `X-RateLimit-Reset` headers on responses. It also caps rows per /enrich request: up to 10,000 with an enrichment_id, and 1,000 with an inline configuration.
Is there a Riveter MCP server?
Yes - Riveter offers an official MCP server. You can also connect the Riveter API directly through Jentic: your agent gets the full 27-endpoint surface without loading another server's tool definitions into its context, your API key is injected at call time by your self-hosted instance instead of living in an MCP config, and your rules decide which operations the agent may call.
How do I enrich a list of companies through Jentic?
Search Jentic for 'enrich a list of companies', load the POST /enrich operation, and execute it with your rows in input plus a prompt and attributes, or an existing enrichment_id. Then load GET /runs/{id}/result to fetch the output. To run it on your own infrastructure, install Jentic One from its GitHub repo.
Can I limit what my agent is allowed to do with the Riveter API?
Yes. For a research agent, write a rule that allows only POST /quick_search, POST /scrape, and GET /runs/{id}/result, so it can search, read pages, and fetch results but cannot start paid enrichments, create monitors, or edit saved enrichments. Every call it makes is logged.
Can I estimate what a Riveter request will cost before running it?
Yes. The kickoff requests in the spec, including /enrich, /datasets, /quick_search, /search_agent, and /scrape, accept an estimate_cost flag and a max_credits ceiling, so an agent can check the cost first and cap how many credits a run may use.
GET STARTED