Wachete

@wachete list all my active watchers and show me which pages they're monitoring

Found 8 active watchers in your account: 'Competitor pricing page' (checks every 6 hours), 'Product changelog' (daily), 'Job postings – Engineering' (every 12 hours), 'API status page' (hourly), 'Terms of Service' (weekly), 'Blog RSS feed' (daily), 'Support KB updates' (every 3 days), and 'Partner directory' (weekly). Each watcher includes the target URL, check frequency, and last-checked timestamp. This illustrative list reflects a typical monitoring setup across pricing intel, content updates, and compliance tracking.

This read-heavy example shows how teams audit their monitoring coverage. Requires a valid API key with read access to watchers. Useful for onboarding new teammates or pruning stale monitors. The MCP returns watcher metadata but not the actual page content or change history.

@wachete create a new watcher for https://example.com/releases monitoring the h2.version-number element, check it daily

Created watcher 'releases monitor' (ID: 7f3a9b2c) targeting https://example.com/releases with XPath selector '//h2[@class="version-number"]'. Check frequency set to daily at 09:00 UTC. The watcher will send notifications when the selected element's content changes. First check scheduled for tomorrow morning. You can view change history in your Wachete dashboard once the initial baseline is captured.

This write example demonstrates setting up automated page monitoring with CSS/XPath selectors. Requires API key with write permissions. Before running, confirm the target URL is publicly accessible and the XPath is valid—incorrect selectors will create a watcher that never triggers. The MCP doesn't validate selector accuracy at creation time.

@wachete pull notifications from the last 7 days and summarize which pages changed most frequently

Retrieved 23 notifications across 8 watchers from the past week. Most active: 'API status page' triggered 9 times (uptime incidents on Jan 15, 17, 19), 'Competitor pricing page' changed 5 times (price adjustments detected), 'Blog RSS feed' added 4 new posts, 'Job postings – Engineering' updated 3 times (new roles posted), and 'Product changelog' changed twice. The API status page's frequency suggests monitoring a volatile endpoint—consider adjusting check frequency or notification thresholds to reduce noise.

This synthesis example pairs Wachete's notification data with AI reasoning to surface patterns. Useful for weekly reviews or identifying over-monitored pages. The time-range filter helps scope analysis. Note that notification volume depends on actual page volatility—a quiet week yields fewer insights. The MCP returns raw notifications; the AI performs the frequency ranking.

When Wachete beats manual checks for pricing intel

A 3-person growth team at a B2B SaaS startup needs to track competitor pricing changes across 8 rival products. They're currently screenshotting pricing pages weekly in a shared doc. Wachete's Create Watcher tool lets them set up XPath monitors on each competitor's pricing table—when a price or tier name changes, the team gets a notification in Slack. This works cleanly if competitors use static HTML pricing tables. The trade-off: if a competitor renders pricing client-side with React or behind a login wall, Wachete can't see it. For public-facing pricing pages updated monthly or quarterly, this MCP turns a 20-minute weekly ritual into a passive alert system. Set it up once, let Switchy run the List Notifications tool in your Monday standup thread.

Why compliance teams use Wachete for SEC filings

A 2-person legal ops team at a fintech monitors SEC EDGAR filings for 15 publicly traded partners. They need to know within 24 hours when a 10-K or 8-K drops. Wachete's XPath monitoring hits the EDGAR RSS feed or specific CIK pages, triggering notifications when new filings appear. The List Watchers and Get Folder Content tools let the team organize monitors by partner entity in Switchy's shared workspace. This beats manual daily checks or paying for a Bloomberg terminal add-on. The boundary: if you need full-text search inside PDFs or structured data extraction from filings, Wachete only catches the page change—you'll still open the PDF manually. For binary 'did something new post' alerts on public regulatory sites, this is the right call.

When Wachete works for niche talent sourcing

A 5-person recruiting agency sources candidates from 12 niche job boards that don't offer RSS or API access. They're manually refreshing boards twice daily to catch new postings before competitors. Wachete's Create Watcher tool monitors each board's listings page via XPath, pinging the team when new <div class='job-card'> elements appear. The Delete Watcher tool cleans up monitors when a client engagement ends. This works if the job boards use server-rendered HTML with stable selectors. The failure mode: boards that lazy-load content or rotate DOM IDs break XPath monitors within weeks. For stable, low-frequency boards posting 2-10 jobs daily, Wachete turns sourcing into a notification queue. If boards update selectors monthly or use heavy JavaScript, budget time for monitor maintenance.