tavily-monitor — Claude Code skill: change monitoring and setup
The tavily-monitor Claude Code skill sets up recurring searches or page checks and stores a baseline. The agent uses it to track new search results or changes to known pages over time.
MIT 2,400 installs ★ 497 updated 2026-10-23
npx skills add tavily-ai/skills --skill tavily-monitor -a claude-codeWhat tavily-monitor does
The skill configures a search or page check to run on a cadence, then stores the first result as a baseline. Search monitoring flags URLs absent from the previous snapshot; extraction monitoring flags changes to a page’s extracted content. It creates monitors/{name}/ with monitor.md and snapshot.json.
The cadence follows the request. If none is given, news and changelogs use a daily schedule, while slower-changing pages use a weekly one. Search runs pass the interval as max_age_hours; extraction runs do not use that parameter. The skill relies on tavily-search. If that skill is missing, its instructions install it and read its SKILL.md. Changing a query or URL list calls for a separate folder and a fresh baseline.
What it helps with
- track new search results over recurring runs
- detect edits to selected pages
- keep a baseline for later comparisons
- set an alert rule for scheduled checks
Example prompts for the agent
- Track new articles about React changes with a daily search.
- Check the pricing page weekly and report changes to its content.
- Set up a recurring check of this changelog and save the first snapshot.
How to install tavily-monitor
1Skills CLI (Vercel)
npx skills add tavily-ai/skills --skill tavily-monitor -a claude-code2Manually: put the files in place
~/.claude/skills/tavily-monitor/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/tavily-monitor/
1Skills CLI (Vercel)
npx skills add tavily-ai/skills --skill tavily-monitor -a codex2Manually: put the files in place
.agents/skills/tavily-monitor/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add tavily-ai/skills --skill tavily-monitor -a cursor2Manually: put the files in place
.cursor/skills/tavily-monitor/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install tavily-ai/skills skills/tasks/tavily-monitor2Manually: put the files in place
.github/skills/tavily-monitor/SKILL.md1Skills CLI (Vercel)
npx skills add tavily-ai/skills --skill tavily-monitor -a gemini-cli2Platform CLI
gemini skills install https://github.com/tavily-ai/skills --path skills/tasks/tavily-monitorManually: put the files in place
.devin/skills/tavily-monitor/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add tavily-ai/skills --skill tavily-monitor -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about tavily-monitor
How do I install tavily-monitor in Claude Code?
Run npx skills add tavily-ai/skills --skill tavily-monitor -a claude-code. The skill file is at ~/.claude/skills/tavily-monitor/SKILL.md.
What does tavily-monitor do?
It configures a recurring search or check of specified pages, runs it once, and saves the result as a baseline. Later runs report changes that match the configured alert rule.
How does tavily-monitor distinguish a search from a page check?
A search flags URLs that were not in the previous snapshot. A page check flags changes to the extracted content of known URLs.
Source text
SKILL.md · MIT
---
name: tavily-monitor
description: |
Build a Tavily search or extraction that runs on a cadence, with a saved baseline. Search monitors set max_age_hours to the interval so a cached copy from before it is not returned. Use when the user wants to monitor, track, watch, or be alerted when something changes, or asks for a recurring check of a query or a page. For a deep comprehensive report use tavily-research. For a one-time lookup use tavily-search. For a one-time table use tavily-list.
allowed-tools: Bash(tvly *)
---
# Tavily monitor
Turn a watch into a search or an extraction that can run again unchanged, then
run it once and save the baseline. The monitor file is the result.
Use `tavily-search` to run the check. If it isn't installed, run `npx skills add tavily-ai/skills --skill tavily-search` and read its SKILL.md before continuing.
## When this is the wrong skill
- The user wants a deep comprehensive report. Use `tavily-research`. If it isn't installed, run `npx skills add tavily-ai/skills --skill tavily-research` and read its SKILL.md before continuing.
- The user wants every current match, once. Use `tavily-list`. If it isn't installed, run `npx skills add tavily-ai/skills --skill tavily-list` and read its SKILL.md before continuing.
## Pick the kind
| You have | Kind | What "changed" means |
|---|---|---|
| A topic, and the pages are not known yet | search | A result URL that was not in the previous snapshot |
| One or more known URLs | extract | The hash of that page's extracted content changed |
Use search for mentions, news, jobs, and anything where new pages appear. Use
extract for a pricing page, a changelog, a docs page, or a status page.
## Steps
1. Write the query or the URL list so the next run can reuse it verbatim. Follow
`tavily-search` for how to shape it. Leave dates and "latest" out of the
query. The snapshot, not the wording, decides what is new.
2. Write the change rule in one sentence. For a search, a new URL is the
alert. A URL from the last snapshot that is missing this run is not an
alert: a page that could not be fetched live is left out. For an extract, a
changed hash is the alert, and the alert quotes the part the user cares
about (the price, the version), not the whole page.
3. Name a cadence. Use the one the user asked for. If they did not, use daily
for news and changelogs, and weekly for a page that changes slowly, and say
which you chose. On a search, set `max_age_hours` to that interval in hours,
so a cached copy older than the cadence is not returned. Daily is 24, weekly
is 168. Use `0` when the interval is under an hour, and `720` when it is
longer than 30 days. Read how the parameter works in the search reference of
`tavily-search`. Write the number in the monitor file and send it on every
run, including the first. Extract does not take `max_age_hours`.
4. Run it once. If the results are the wrong kind of thing, fix the query or
the URLs and run again before you save anything. If the search is refused
with `434`, the run did not complete. Say so, and leave `snapshot.json` as
it was.
5. Save the monitor and the snapshot. The next run reads the monitor, runs the
same search or extract, reports only what the change rule flags, and
replaces `snapshot.json`.
Create `monitors/{name}/` in the working directory. The name is sentence case,
3 to 6 words, letters, numbers, and single spaces. Quote it in shell commands.
Do not start a scheduler unless the user asks for one. When they do, and the
environment has a way to run on that cadence, point it at this folder.
A later change to the query or the URL list is a new monitor. It gets a new
folder and a new baseline. Editing them in place makes the next diff meaningless.
## Output
`monitors/{name}/monitor.md`:
```markdown
# {Name}
- Cadence: {daily | weekly on Monday | the schedule the user asked for}
- Kind: search | extract
- Query: {the query, for a search}
- URLs: {one per line, for an extract}
- max_age_hours: {the cadence in hours, search only}
- Alert when: {the change rule}
- Snapshot: snapshot.json
## Run again
Read this file. Use `tavily-search` to run the same search or extract. On a
search, pass `max_age_hours` from this file. Compare the result to
snapshot.json using the alert rule above. Report only the changes. Then
replace snapshot.json with this run. If the search is refused with `434`,
report that and keep the snapshot.
```
`monitors/{name}/snapshot.json`:
```json
{
"ran_at": "YYYY-MM-DD",
"kind": "search",
"query": "",
"items": [{ "url": "", "title": "", "snippet": "" }]
}
```
For an extract, store `pages` instead of `items`: `url`, `sha256` of the
extracted content, and a short excerpt of the part being watched.
In the reply, give the cadence, `max_age_hours` for a search, the query or the
URLs, the alert rule, and the folder path, plus what the baseline already shows.
© tavily-ai · MIT · tavily-ai/skills