Regulatory Feed Monitor (Federal Register + SEC)

Scheduled monitoring of the U.S. Federal Register (any agency, by slug) plus SEC press-release & litigation RSS — and any other RSS/Atom feed you add — for compliance teams and AI agents. It finds newly published rules, notices, and releases matching your agencies + keywords and emits clean, structured, RAG-ready records.

Built by a branch compliance manager to cut down the manual work of watching for newly published regulatory items. (Tracks new items, not edits to existing documents — see Limitations.)

What it actually does

• Queries the Federal Register JSON API (free, official, keyless) filtered by the agency slugs and keywords you provide.
• Parses any RSS/Atom feeds you list. Defaults: two SEC feeds (press releases + litigation releases).
• Flags items it hasn't seen before across runs (ID-based), so a scheduled run emits only new entries.
• Adds a lightweight inferred compliance area and a one-line suggested action to each item.

What you get (per item)

field	meaning
title, summary, url	the document (summary = Federal Register abstract or RSS summary)
source, agency, type	Federal Register or the RSS feed title, and document type
published_date, retrieved_at	timing
keywords_matched	which of your terms appeared in the title/abstract
affected_area	a keyword-heuristic guess (Reg BI, AML, Reg S-P, Advertising, Custody…); often "General / uncategorized"
action_required	a one-line heuristic prompt (update WSPs / review / watch enforcement) based on document type
markdown	included only when outputFormat: "markdown"

Input

• agencies — Federal Register agency slugs (default: securities-and-exchange-commission, commodity-futures-trading-commission). To cover CFPB, FinCEN, CFTC, etc., add their Federal Register slugs and/or their RSS feeds.
• keywords — filter to matching items. Federal Register results are filtered server-side by full-text term search (one query per keyword); RSS items are kept only if a keyword appears as a substring of the title/summary. Leave empty to get everything from the selected agencies/feeds.
• daysBack — Federal Register lookback window (default 7).
• rssFeeds — extra feeds (SEC/FINRA/state). Defaults to two SEC feeds.
• onlyNew — remember IDs across runs and emit only unseen items (default true).
• outputFormat — structured (default) or markdown.

Use it as a scheduled monitor

Set an Apify Schedule (e.g., daily) with onlyNew: true. Each run pushes only newly seen items to the dataset — pipe that into Slack, email, a list, or a RAG knowledge base.

Limitations — read this

• Not real-time. It polls on each run; it does not stream or push. Freshness = your schedule + the Federal Register's own publishing lag.
• New-item detection only, not content-change detection. Tracking is by document ID. If a document is revised but keeps its ID/number, it will not be re-flagged. "Change" here means "new item since last run."
• No built-in CFPB / FinCEN / CFTC connectors. Those are reachable only if they publish in the Federal Register (add the slug) or you supply their RSS feed. Only the Federal Register and your RSS feeds are queried.
• affected_area and action_required are simple keyword / document-type heuristics, not legal analysis. They miss often and can mislabel. Treat them as triage hints.
• Keyword matching is literal (case-insensitive) — no synonyms or semantic search. For RSS it is substring matching on title/summary; for the Federal Register it is the API's own full-text term search, so the keywords_matched field (which only checks the title/abstract substring) can be empty even on a returned item that matched on body text.
• RSS coverage depends on the feed. Default SEC feeds may change or rate-limit; verify feeds for your use.
• Not legal or compliance advice.

Sources

• Federal Register API: https://www.federalregister.gov/developers/documentation/api/v1
• SEC press releases & litigation releases RSS (default feeds; configurable).