ceven

Perigon

Pulls structured news data and web content into your workflows to track competitors, monitor brand mentions, and automate industry research reports.

Try Perigon in Ceven

Ask Ceven anything
Standard

Why use Ceven?

  1. AI native Perigon integration

    • Describe the outcome and Ceven picks the right Perigon calls, fills the parameters, and checks the result.
    • Structured, agent friendly tool schemas so each call runs reliably instead of by guesswork.
    • Rich coverage for reading, writing, and querying your Perigon data, across all 9 of its actions.

  2. Managed auth

    • Built in OAuth with automatic token refresh and rotation.
    • One place to manage, scope, and revoke Perigon access.
    • Per user and per environment credentials instead of shared keys.

  3. Agent optimized design

    • Actions are tuned from real success and error rates so reliability climbs over time.
    • Full execution logs so you always know what ran in Perigon, when, and on whose behalf.
    • The agent pauses and asks when Perigon is unclear instead of plowing ahead.

  4. Enterprise grade security

    • Fine grained access so you control which agents and people can reach Perigon.
    • Least privilege by default, read scopes first and only the writes a workflow needs.
    • A full audit trail of every Perigon action to support review and sign off.

Supported tools

Every action Ceven's agents can run on Perigon, and when to use it.

Search articles
Use this to find news articles based on keywords, dates, or specific publishers. Ideal for gathering historical context on a company.
Get article content
Pull the full body text and metadata for a specific article ID to use in summaries or sentiment analysis.
List recent news
Pull a stream of the latest articles across the entire index. Use this to build a general industry ticker.
Filter by entity
Retrieve all articles mentioning a specific person or organization. Use this for brand monitoring.
Get source metadata
Pull details about a specific news publisher to verify credibility or bias before including it in a report.
Search by domain
Limit news searches to a specific set of trusted domains or websites to remove noise.
Get article metadata
Pull only the headline, author, and date without the full body. Use this for high level indexing.
Track keyword
Set up a monitor for a specific term. The agent uses this to trigger downstream workflows when the term appears.
Get content by URL
Pass a specific web link to Perigon to get a structured version of that page for analysis.
List categories
Pull the available news categories to help the agent narrow down searches by topic.
Get trending topics
Pull a list of currently spiking keywords across the web to identify emerging trends.
Validate API status
Check the current health of the Perigon API to ensure news feeds are flowing correctly.

12 actions · scroll to see them all

Frequently asked questions

Ceven manages Perigon rate limits through an internal queuing system that respects the specific tier of your API key. If a workflow requests a massive bulk pull of historical articles, the agent will automatically paginate the requests and introduce small delays to prevent 429 errors. You can see the current status of these requests in the workflow logs. If you consistently hit limits, the agent will suggest upgrading your Perigon plan to increase your requests per minute. We prioritize the delivery of real time alerts over historical backfills to ensure your most critical news triggers always fire without delay.
Ceven uses the source metadata provided by Perigon to apply custom quality filters. You can define a whitelist of trusted domains or a blacklist of known low quality sites within your workflow settings. The agent can also run a secondary check using an LLM to score the credibility of an article based on the publisher and tone before it ever hits your Slack or CRM. This ensures that only high signal information triggers your business processes. You can adjust these sensitivity thresholds in the agent configuration panel at any time to be more inclusive or more restrictive.
Perigon has different archival depths depending on the data source and your specific subscription tier. While many global news sources are indexed for several years, some niche blogs or regional sites may only have data going back a few months. When you run a historical search, Ceven will inform you of the earliest date available for the results returned. If you need data older than what the API provides, the agent can attempt to find archived versions via other tools if you have them connected, though the structured format of Perigon is always the primary preference for consistency.
Perigon provides structured data for content it can legally access and index. If an article is behind a hard paywall that prevents indexing, Perigon may only provide the snippet or the metadata rather than the full body text. In these cases, Ceven will notify you that the full text is unavailable and provide the direct link to the source. You can configure your workflow to handle these gaps by either ignoring them or flagging them for a human to manually purchase access and upload the text for the agent to process.
Perigon focuses primarily on structured news and web content rather than raw social media streams. While some social media posts that are picked up by news aggregators may appear in the results, it is not a replacement for a dedicated social listening tool. Use Perigon for articles, press releases, and official web content. If your workflow requires tracking individual tweets or LinkedIn posts, we recommend connecting a dedicated social API alongside Perigon. The agent can then synthesize data from both sources into a single consolidated report for your team.
The index is updated in near real time, but there is a slight propagation delay between a story being published and it appearing in the API. This delay varies by source; major news wires are indexed almost instantly, while smaller blogs might take several minutes or hours. Ceven handles this by polling the API at intervals you define in your workflow. For mission critical alerts, we recommend a five minute polling window to balance timeliness with API efficiency. The agent always includes the original publication timestamp so you know exactly how fresh the information is.
Perigon provides the clean structured text, but it does not provide its own sentiment scoring. Instead, Ceven passes the structured text from Perigon into a high performance LLM to perform the sentiment analysis. This is actually an advantage because you can tell the agent exactly what to look for. For example, instead of a generic positive or negative score, you can ask the agent to score the text based on bullish or bearish sentiment specifically for a stock ticker, or to identify specific risks like regulatory hurdles or leadership changes.
If you use the get content by URL action and the page is not in the Perigon index or is inaccessible, the API will return an error. Ceven is programmed to handle this gracefully by attempting a secondary fetch using a basic web scraper if enabled. If both methods fail, the agent will mark that specific item as failed in your workflow and move to the next one, providing a summary of the missing links at the end. This prevents a single dead link from crashing a large scale research automation.

Alternatives to Perigon

Other tools that solve a similar problem. Ceven supports these too, so you can switch or run more than one at once.

NewsAPI logoNewsAPIEvent Registry logoEvent RegistryContextual Web Search logoContextual Web Search

Try Ceven on your stack

Plug Ceven on top of the tools you already run. Connect Perigon and the rest of your stack, describe the outcome, and its agents handle the work end to end, days of it in minutes.

Get started for free