Reg Feed Watcher

/reg-feed-watcher

1. Load ~/.claude/plugins/config/claude-for-legal/regulatory-legal/CLAUDE.md → watchlist, materiality threshold, feed config.
2. Use the workflow below.
3. Pull each feed. Filter by materiality.
4. Output: what's new, categorized by materiality tier.

Purpose

Pull the feeds. Filter by materiality. Output what's left. The filter is the value — unfiltered feeds are noise.

Load context

~/.claude/plugins/config/claude-for-legal/regulatory-legal/CLAUDE.md → watchlist, materiality threshold, feed configuration, digest output path (if set).

references/source-catalog.md (in this skill's directory) → curated catalog of RSS/JSON/HTML sources across US federal, US state, EU/UK, international, and secondary/aggregator categories. Use when configuring new sources or when the user's watchlist has coverage gaps (see Step 0).

Workflow

Step 0: Coverage check (before pulling)

Before running the pull, compare the watchlist + feed configuration in CLAUDE.md against references/source-catalog.md:

• Which categories (US federal / US state / EU-UK / international) does the user care about per their watchlist?
• Which of those categories have zero or very few sources configured?

If there's an obvious gap — e.g., user watches "EU regulators" in the watchlist but has only edpb.europa.eu configured in feeds, missing ICO, CNIL, DPC Ireland — surface it once at the top of the digest:

Coverage gap noticed: Your watchlist includes [category], but only [N] feeds are configured. The source catalog lists [X] options in this category (e.g., [top 2-3 names]). Want me to suggest additions? Run /regulatory-legal:cold-start-interview --redo to update, or edit ~/.claude/plugins/config/claude-for-legal/regulatory-legal/CLAUDE.md directly.

Don't nag the same gap repeatedly — if the user has explicitly said "skip state AGs for now," respect that. Note state in CLAUDE.md so it sticks.

Step 1: Pull

Pull from all configured feed tiers. Every installation has Tier 1. Tiers 2 and 3 are additive — use them if configured, skip if not.

Tier 1 — Free feeds (always active)

For each regulator in the watchlist:

• Federal Register API (https://www.federalregister.gov/api/v1/documents) — query by agency slug, date range (since last check), document type. Returns structured data: document type, title, abstract, effective date, comment deadline (for NPRMs), and citation. Covers all US federal agencies.
• Direct regulator RSS — fetch and parse any RSS URLs in ~/.claude/plugins/config/claude-for-legal/regulatory-legal/CLAUDE.md feed configuration (SEC, FTC, CFPB, state agencies, EU DPAs, etc.).

Agency slug reference for common watchlist regulators: | Regulator | API slug | |---|---| | FTC | federal-trade-commission | | SEC | securities-and-exchange-commission | | CFPB | consumer-financial-protection-bureau | | CPPA (CA) | RSS only — cppa.ca.gov/feed | | DOL | labor-department | | HHS | health-and-human-services-department | | FCC | federal-communications-commission |

For any regulator not in this list: check federalregister.gov/agencies for the correct slug, or fall back to direct RSS.

Tier 2 — Paid feeds (if configured)

• Paid regulatory feed MCP: Query for updates since last check date, filtered to watchlist regulators.
• CourtListener MCP: Same.

De-duplicate across tiers — the same document may appear in multiple sources. Prefer the richest source for the enriched output.

No silent supplement. If the feed pull returns few or no results for a regulator in the watchlist, report what was found and stop. Do NOT fill the gap from web search or model knowledge without asking. Say: "The feed check returned [N] items from [regulators hit]. Coverage appears thin for [regulator / topic]. Options: (1) broaden the date window, (2) try a different feed or MCP, (3) search the web — results will be tagged [web search — verify] and should be checked against the issuing authority's website before relying, or (4) stop here. Which would you like?" A lawyer decides whether to accept lower-confidence sources; Claude does not decide for them.

Source attribution. Tag every citation and regulatory item with where it came from: [Federal Register], [<regulator> RSS], [CourtListener], or the specific MCP tool name for items retrieved via connector; [web search — verify] for items from web search; [model knowledge — verify] for items surfaced from the model's training data; [user provided] for manually-pasted items. Items tagged verify carry higher fabrication risk than tool-retrieved items and should be checked first. Never strip or collapse the tags — they are the user's fastest signal about which citations to verify.

Secondary sources. Some catalog entries (IAPP, FPF, Hogan Lovells, Covington, Lexology, JD Supra, Artificial Lawyer, LawSites, and similar commentators/aggregators) report on primary regulatory action but are not the primary source. Tag any item pulled from these feeds with [secondary source] in addition to the feed-name tag — e.g., [IAPP Daily Dashboard] [secondary source]. In the digest, when a secondary-source item describes a regulator action, add a note: "→ Trace to primary: [link to regulator site if known, otherwise 'find on <regulator>.gov before relying']." Do not classify a secondary-source item as "Always material" on its own strength — bump it down a tier until the primary source is located.

Tier 3 — Manual entry

If the user has pasted regulatory text or a summary rather than invoking from a scheduled feed check: treat the pasted content as a single item, skip to Step 2 for classification, and record source as "manual entry." No feed pull required. This path works regardless of subscription status.

Record the check timestamp after pulling. Next scheduled run pulls from here forward.

Step 2: Classify

Each item gets a materiality tier per ~/.claude/plugins/config/claude-for-legal/regulatory-legal/CLAUDE.md:

ANPR / RFI handling — specific. Pre-rule items are distinct from NPRMs in one important way: they don't change the law, but they do carry comment deadlines and they signal the regulator's direction. Treat them as a separate branch:

• Do not classify an ANPR / RFI as "always material" — the compliance impact is zero until a rule issues.
• Do classify as review-worthy if any of the issue areas in the notice touch the watchlist's always-material categories (e.g., an ANPR on open banking in a fintech watchlist).
• Do log the comment deadline to ~/.claude/plugins/config/claude-for-legal/regulatory-legal/comment-tracker.yaml with item_type: ANPR or item_type: RFI so the downstream tracker can distinguish these from compliance gaps.