Site Spy
NPM · @SITE-SPY/MCP-SERVER · SCANNED SEP 25
MCP server for Site Spy — lets AI agents monitor websites for changes
Available components
How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. How we score → Why this is hard to score →
Supply Chain Security98
- No malware found by supply-chain analysis.Pass
- No known CVEs affecting this package version or its production dependencies.Pass
- No install/post-install scripts declared.Pass
- 31 of 95 dependencies flagged as unhealthy. View diagnostics → Partial
Provenance & Transparency19
- Repository check failed: no source repository is declared. See how to fix → View diagnostics → Fail
- Provenance check failed: no build-provenance attestation is published. See how to fix → View diagnostics → Fail
- Clear OSI-approved license (ISC).Pass
- Actively maintained (last published 2 days ago).Pass
- Security-disclosure policy not yet verified: we couldn't inspect the source repository.Unverified
Schema Quality & AI Usability80
- AI-judged instruction clarity (excellent).Pass
- Tool/resource definitions use about 1714 tokens (~90/item across 19 items; 19 tools + 0 resources), lean.Pass
- Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management85
- Stability check failed: the tool surface changed between 0.2.2 and 0.3.1: 1 tool removals, 0 breaking changes, 0 additions. See how to fix → Fail
Tool Coverage100
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 100% of tool parameters carry a description.Pass
Tool Safety100
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- All 2 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.Pass
- An AI judge read all 19 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
- Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
How do I install the Site Spy MCP server?
Site Spy runs locally as an npm package, launched with npx -y @site-spy/mcp-server. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.
npm · @site-spy/mcp-server
claude mcp add vkuprin-site-spy -- npx -y @site-spy/mcp-server
{
"mcpServers": {
"vkuprin-site-spy": {
"command": "npx",
"args": [
"-y",
"@site-spy/mcp-server"
]
}
}
} {
"servers": {
"vkuprin-site-spy": {
"command": "npx",
"args": [
"-y",
"@site-spy/mcp-server"
]
}
}
} codex mcp add vkuprin-site-spy -- npx -y @site-spy/mcp-server
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"vkuprin-site-spy": {
"type": "local",
"command": [
"npx",
"-y",
"@site-spy/mcp-server"
],
"enabled": true
}
}
} openclaw mcp add vkuprin-site-spy --command npx --arg -y --arg @site-spy/mcp-server
mcp_servers:
vkuprin-site-spy:
command: "npx"
args: ["-y", "@site-spy/mcp-server"] {
"McpServers": {
"vkuprin-site-spy": {
"Transport": "stdio",
"Command": "npx",
"Arguments": [
"-y",
"@site-spy/mcp-server"
]
}
}
} assistant mcp add vkuprin-site-spy -t stdio -c npx -a -y @site-spy/mcp-server
{
"mcpServers": {
"vkuprin-site-spy": {
"command": "npx",
"args": [
"-y",
"@site-spy/mcp-server"
]
}
}
} Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.
- 25 Sept 26 +1
- We updated how we score, so this day's move reflects our rubric, not a change to the server See what changed → functional
- 23 Sept 26 −2
No change was recorded against any check on this day. Stability & Change Management went from 95 to 78.
- 22 Sept 26 +1
- Stability: fail → unverified ▼ security
- Tool safety: pass → unverified ▼ security
- Capabilities: pass → unverified ▼ functional
- Tool coverage: 100 → unverified ▼ functional
- Package version: 0.3.0 → 0.3.1 functional
- 19 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 82 to 85.
- 17 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 75 to 78.
- 16 Sept 26 −3
No change was recorded against any check on this day. Stability & Change Management went from 95 to 75.
- 15 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 92 to 95.
- 12 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 82 to 85.
Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.
Captured 25 Sept 2026 · Analysed npm/@site-spy/mcp-server@0.3.1
Provenance No attestation
The registry publishes no build provenance for this version, so there is nothing to verify.
| Result | No attestation |
|---|---|
| Ecosystem | npm |
Background: How many MCP packages publish verified provenance →
Dependencies 95 packages
| Packages resolved | 95 |
|---|---|
| Stale | 31 |
| Tree resolution | Complete |
Background: SBOMs and build attestations, explained →
The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →
auth_status Check connection ~20
Check whether the MCP server is currently authenticated with Site Spy.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
authenticate Connect to Site Spy ~85
Connect to Site Spy with an API key. If the user doesn't have one, ask them to open https://sitespy.app/dashboard to log in and copy their key from the Settings > API tab. The key is held in memory only and is lost when the server restarts — set SITE_SPY_API_KEY to persist it.
| Name | Type | Req | Description |
|---|---|---|---|
| api_key | string | yes | Site Spy API key |
No output schema declared.
No examples provided.
create_watch Create watch ~151
Start monitoring a URL for changes. Returns the UUID of the new watch.
| Name | Type | Req | Description |
|---|---|---|---|
| check_hours | number | – | Hours between checks (default: uses account setting) |
| check_minutes | number | – | Minutes between checks |
| fetch_backend | string | – | How to fetch the page: html_requests (fast, no JS), html_webdriver (browser with JS, for SPAs), or html_hybrid (plain HTTP first, then a browser and the anti-bot solvers when the page needs them). Pr… |
| title | string | – | Human-readable name for this monitor |
| url | string | yes | URL to monitor for changes |
No output schema declared.
No examples provided.
delete_watch Delete watch ~32
Stop monitoring a URL and delete the watch with all its history.
| Name | Type | Req | Description |
|---|---|---|---|
| uuid | string | yes | Watch UUID to delete |
No output schema declared.
No examples provided.
fetch_url Fetch a URL ~470
Fetch a single known URL through Site Spy. Do NOT use this for discovery — use web_search to find URLs first, then fetch_url on each result. Do NOT fetch google.com / bing.com directly (they always block — wastes the cascade budget). For known static sites prefer fetch_backend: 'html_requests' (≤10s). For SPAs use 'html_webdriver'. 'auto' starts cheap and escalates only when the site blocks us — plain HTTP, then a browser-grade TLS fingerprint, then a browser, then the anti-bot solvers — capped at 50s total (FETCH_URL_AUTO_TOTAL_TIMEOUT_SECONDS). On failure, response includes attempts[] and cascade_summary so you can see what was tried. This does not create a watch or save history.
| Name | Type | Req | Description |
|---|---|---|---|
| body | string | – | Optional request body string. |
| fetch_backend | string | – | Fetch strategy. Default 'auto' escalates on a block: html_requests -> impersonate (Chrome TLS, no browser) -> html_webdriver -> flaresolverr -> stealth_browser, with a 50s total cap. Prefer 'html_req… |
| headers | object | – | Optional request headers. |
| ignore_status_codes | boolean | – | Return content for non-2xx responses instead of failing. Defaults to true. |
| include_filters | array | – | Optional selectors used by auto mode to decide if browser rendering is needed. |
| js | string | – | JavaScript to execute before extraction when a browser backend is used. |
| max_content_chars | integer | – | Maximum returned characters for content fields. |
| method | string | – | HTTP method. Browser fetchers use GET-like navigation. |
| output | string | – | Response content format. Defaults to text. |
| screenshot | boolean | – | Return a base64 screenshot when the browser backend captures one. |
| screenshot_format | string | – | Screenshot format. Defaults to PNG. |
| timeout | integer | – | Per-backend timeout in seconds (1-120). 'auto' mode also has a separate 30s total wall-clock cap (env FETCH_URL_AUTO_TOTAL_TIMEOUT_SECONDS) that bounds the whole cascade. |
| url | string | yes | URL to fetch |
| wait_seconds | integer | – | Extra browser wait before extracting content. |
No output schema declared.
No examples provided.
generate_rss_token Generate RSS token ~76
Generate or regenerate an RSS feed token. Use the token to build feed URLs: {server}/rss?token={token} for all watches, {server}/rss/watch/{uuid}?token={token} for a single watch. Regenerating invalidates all existing feed URLs, so any reader already subscribed will stop receiving updates.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
get_ai_summary Get AI change summary ~56
Get the AI-generated plain-English summary of a watch's most recent change. Requires a Pro or Business plan. Returns the summary text if available, or a message if no summary exists yet.
| Name | Type | Req | Description |
|---|---|---|---|
| uuid | string | yes | Watch UUID |
No output schema declared.
No examples provided.
get_change_history List change timestamps ~77
Get the list of timestamps when changes were detected for a watch. Use these timestamps with get_snapshot or get_diff.
| Name | Type | Req | Description |
|---|---|---|---|
| limit | integer | – | Return at most this many entries. Default: all of them. |
| offset | integer | – | Skip this many entries before returning. Use with limit. |
| uuid | string | yes | Watch UUID |
No output schema declared.
No examples provided.
get_diff Compare snapshots ~98
Compare page content between two timestamps to see exactly what changed. Use get_change_history to find valid timestamps. Output is capped at 20000 characters by default — raise max_chars if you need more.
| Name | Type | Req | Description |
|---|---|---|---|
| from_timestamp | string | yes | Earlier timestamp |
| max_chars | integer | – | Maximum characters of diff output to return. Defaults to 20000. |
| to_timestamp | string | yes | Later timestamp |
| uuid | string | yes | Watch UUID |
No output schema declared.
No examples provided.
get_notifications Get notification settings ~23
Get the current notification settings (email, Telegram, web push, etc.).
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
get_rss_settings Get RSS settings ~26
Get the current RSS feed settings including whether feeds are enabled and the feed token.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
get_snapshot Get snapshot ~88
Get the page content captured at a specific timestamp. Use get_change_history first to find valid timestamps. Content is capped at 20000 characters by default — raise max_chars if you need more.
| Name | Type | Req | Description |
|---|---|---|---|
| max_chars | integer | – | Maximum characters of page content to return. Defaults to 20000. |
| timestamp | string | yes | Timestamp from change history |
| uuid | string | yes | Watch UUID |
No output schema declared.
No examples provided.
get_watch Get watch ~37
Get full details of a specific watch including its configuration, last check status, and change history summary.
| Name | Type | Req | Description |
|---|---|---|---|
| uuid | string | yes | Watch UUID |
No output schema declared.
No examples provided.
list_watches List watches ~74
List all monitored websites. Returns an object keyed by watch UUID, each with its URL, title, status, and last checked/changed timestamps.
| Name | Type | Req | Description |
|---|---|---|---|
| limit | integer | – | Return at most this many watches. Default: all of them. |
| offset | integer | – | Skip this many watches before returning. Use with limit. |
No output schema declared.
No examples provided.
revoke_rss_token Revoke RSS token ~24
Revoke the RSS feed token, disabling all RSS feed access immediately.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
search_watches Search watches ~40
Search across all watches by URL, title, or content. Returns matching watches.
| Name | Type | Req | Description |
|---|---|---|---|
| query | string | yes | Search query (matches against URL, title) |
No output schema declared.
No examples provided.
trigger_recheck Recheck all watches ~44
Force an immediate recheck of every watch. Useful when you need fresh data now rather than waiting for the next scheduled check. Manual rechecks are capped per day by plan.
Input schema present but exposes no named parameters.
No output schema declared.
No examples provided.
update_watch Update watch ~110
Update a watch's configuration. Use this to pause/resume, change check interval, update title, or modify notification settings.
| Name | Type | Req | Description |
|---|---|---|---|
| check_hours | number | – | Hours between checks |
| check_minutes | number | – | Minutes between checks |
| notification_muted | boolean | – | Mute or unmute notifications |
| paused | boolean | – | Pause or resume monitoring |
| title | string | – | New title |
| url | string | – | New URL to monitor |
| uuid | string | yes | Watch UUID |
No output schema declared.
No examples provided.
web_search Search the web ~183
Search the web via DuckDuckGo HTML and return a list of result URLs with titles and snippets. Use this BEFORE fetch_url when the user asks you to 'search', 'find', 'look up', or otherwise discover content — instead of guessing google.com (which always blocks). Pipeline: web_search(query) -> pick relevant results -> fetch_url(result.url, fetch_backend: 'html_requests') on each. Returns up to 20 results.
| Name | Type | Req | Description |
|---|---|---|---|
| debug | boolean | – | When true, include a debug_html_excerpt in the response for troubleshooting parsing failures. |
| limit | integer | – | Max results to return. Default 10, max 20. |
| query | string | yes | Search query. |
| region | string | – | DuckDuckGo region code. Default 'wt-wt' (worldwide). Examples: 'us-en', 'fr-fr', 'de-de'. |
No output schema declared.
No examples provided.
What is the Site Spy MCP server?
Site Spy is an MCP server listed in the public MCP registry as io.github.vkuprin/site-spy. MCP server for Site Spy, lets AI agents monitor websites for changes. This page covers its npm package (@site-spy/mcp-server).
Is the Site Spy MCP server safe to use?
Site Spy scores 78 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 25 September 2026. It declares no install or post-install scripts. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.
What tools does the Site Spy MCP server expose?
Site Spy exposes 19 tools: authenticate, auth_status, list_watches, create_watch, get_watch, and 14 more. Their descriptions and schemas cost roughly 1,714 tokens of context every time the server is loaded.
Is the Site Spy MCP server still maintained?
Site Spy is still listed as active in the MCP registry. We last reached this channel on 25 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.
What licence is the Site Spy MCP server under?
Site Spy declares the ISC licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.