# AISeeWhy (remote · api.aiseewhy.com)

Can ChatGPT, Claude and Perplexity crawlers read your page? Audit, fixes, compare, monitoring.

- Trust score: 77/100 (medium)
- Change this week: +3
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-21

## Components

- remote · `api.aiseewhy.com`: 77/100 (this document), [markdown](https://verifymcp.io/servers/com-aiseewhy-aiseewhy/api.md), [page](https://verifymcp.io/servers/com-aiseewhy-aiseewhy/api)

## Channel facts

- Endpoint: `https://api.aiseewhy.com/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.2.0`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-21.

- **Endpoint Security**: 80/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - No authorisation is required to call this server. Nothing on it is marked destructive, but we can't show that an anonymous caller can only read: full credit needs the server to list at least one tool and every tool to be verifiably read-only. We use a best-effort check of tool names and descriptions to tell read-only tools from tools that change things, so some genuinely read-only tools may not get full credit. Names that start with a read verb (get, list, read, search and similar) and accurate readOnlyHint annotations help. Meet that to earn full credit, or require authorisation.
  - HTTPS is enforced; there's no plaintext access path.
  - The HSTS (Strict-Transport-Security) header is present.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 79/100
  - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 3060 tokens (~153/item across 20 items; 19 tools + 1 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 30/100
  - Stability observed for 9 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
  - Structured output schemas are declared (100% of tools); any adoption earns full credit.
- **Tool Safety**: 75/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - 0 of 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "generate_content_merge_plan" implies "merge" and declares readOnlyHint instead, contradicting what its own name says it does.
  - An AI judge read all 21 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a current MCP spec version (2026-07-28).
  - Supports UI / widget rendering.

## Install

### How do I install the AISeeWhy MCP server?

AISeeWhy is a hosted endpoint at https://api.aiseewhy.com/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http com-aiseewhy-aiseewhy 'https://api.aiseewhy.com/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "com-aiseewhy-aiseewhy": {
      "url": "https://api.aiseewhy.com/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "com-aiseewhy-aiseewhy": {
      "type": "http",
      "url": "https://api.aiseewhy.com/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.com-aiseewhy-aiseewhy]
url = "https://api.aiseewhy.com/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "com-aiseewhy-aiseewhy": {
      "type": "remote",
      "url": "https://api.aiseewhy.com/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add com-aiseewhy-aiseewhy --url 'https://api.aiseewhy.com/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  com-aiseewhy-aiseewhy:
    url: "https://api.aiseewhy.com/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "com-aiseewhy-aiseewhy": {
      "Transport": "http",
      "Url": "https://api.aiseewhy.com/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add com-aiseewhy-aiseewhy -t streamable-http -u 'https://api.aiseewhy.com/mcp'
```

### Other

```json
{
  "mcpServers": {
    "com-aiseewhy-aiseewhy": {
      "type": "http",
      "url": "https://api.aiseewhy.com/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-21 (score 77, +1)

- [security] The server rewrote its instructions, which are the text every model session reads
- [functional regression] Schema quality: 1253 → 3060
- [functional regression] Schema quality: 1253 → 1452
- [functional] First check of Capabilities: pass
- [functional] First check of Schema quality: 100
- [functional] Destructive annotations: pass → 0
- [functional] New resource “ai-crawler-check”
- [functional] Server version: 1.1.0 → 1.2.0
- [functional] New tool “detect_keyword_cannibalization”
- [functional] New tool “find_content_consolidation_candidates”
- [functional] New tool “find_declining_pages”
- [functional] New tool “find_high_impression_low_ctr_pages”
- [functional] New tool “find_zero_click_pages”
- [functional] New tool “generate_content_merge_plan”
- [functional] New tool “generate_redirect_plan”
- [functional] New tool “get_all_pages_performance”
- [functional] New tool “identify_strongest_page”
- [functional] New tool “prioritize_seo_opportunities”
- [functional] New tool “show_ai_crawler_check”
- [cosmetic] “audit_sitemap” reworded the description of “language”
- [cosmetic] “audit_url” reworded the description of “language”
- [cosmetic] “compare_urls” reworded the description of “language”
- [cosmetic] “crawl_site” reworded the description of “language”
- [cosmetic] “explain_finding” reworded the description of “language”

### 2026-11-20 (score 76, +1)

No change was recorded against any check on this day. Stability & Change Management went from 23 to 27. That category is still filling its 30-day observation window: 7 days of observed history at the previous scan, 8 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-18 (score 75, 0)

- [security] Tool “crawl_site” rewrote its description, which is the text the model reads

### 2026-11-17 (score 75, +1)

No change was recorded against any check on this day. Stability & Change Management went from 13 to 17. That category is still filling its 30-day observation window: 4 days of observed history at the previous scan, 5 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-15 (score 74, 0)

- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “audit_url” rewrote its description, which is the text the model reads
- [functional regression] Schema quality: 131 → 157
- [functional] New tool “crawl_site”
- [functional] New tool “find_broken_links”

### 2026-11-14 (score 74, +1)

No change was recorded against any check on this day. Stability & Change Management went from 3 to 7. That category is still filling its 30-day observation window: 1 days of observed history at the previous scan, 2 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-13 (score 73, 0)

- [functional improvement] Stability: unverified → 0.03

### 2026-11-12 (score 73)

First indexed and scored.

## MCP tools (19)

### `show_ai_crawler_check` (~171 tokens)

AI crawler check card

Check one page for AI crawlers and AI search (the same check and limits as audit_url) and show the result as a visual card in apps that render one (ChatGPT, Claude): the score, which AI crawlers are let in or blocked, the costliest finding with its fix and how many findings are locked, by priority. Use it when the user wants to see a site's AI crawler access at a glance; use audit_url for a Markdown report.

Input parameters:

- `language`: Report language as a code such as en, de, fr, es, ja or zh. Defaults to the account language (English without a key); 60 languages are supported.
- `url` (string, required): Page to audit: a full URL or a bare domain such as example.com (https is assumed).

Output parameters:

- `access` (array): One row per AI crawler: bot, cls, status, verdict.
- `counts`: Findings per priority, P1 to P3, of the whole report.
- `domain` (string): Host of the checked page.
- `full` (boolean): True: every finding is in findings. False: a preview, the rest only by priority.
- `lead`: The costliest finding with its fixes, the one the preview opens.
- `links` (object): report: the page on aiseewhy.com; example: a sample full report.
- `locked_counts` (object): Findings not shown, per priority.
- `note`: Why no new check ran (the site was checked before), or null.
- `score`: value 0-100 and visible: whether AI crawlers and AI search can read the page.
- `tone`: Score ring tone: ok, p2 or p1.

### `audit_url` (~213 tokens)

Audit one URL: fetch it as 16 AI and search crawlers plus curl and a browser, check robots.txt, structured data, answer-first content, freshness, hreflang and llms.txt, plus classic on-page SEO (title, meta description, H1, canonical, noindex, image alt text, broken links). Returns a Markdown report with findings and fixes. language: report language code (en, de, ja, …).

Input parameters:

- `bots`: Only test these crawlers, by name (for example GPTBot, OAI-SearchBot, ClaudeBot, PerplexityBot, Google-Extended). Omit to test all 16 AI and search crawlers.
- `language`: Report language as a code such as en, de, fr, es, ja or zh. Defaults to the account language (English without a key); 60 languages are supported.
- `url` (string, required): Page to audit: a full URL or a bare domain such as example.com (https is assumed).

Output parameters:

- `result` (string)

### `audit_sitemap` (~133 tokens)

Audit up to 10 pages listed in a sitemap or sitemap index. Returns one summary line per page plus the full report of every page with findings. Needs an API key.

Input parameters:

- `language`: Report language as a code such as en, de, fr, es, ja or zh. Defaults to the account language (English without a key); 60 languages are supported.
- `max_urls` (integer): How many page URLs from the sitemap to audit, 1 to 10.
- `sitemap_url` (string, required): URL of a sitemap.xml or sitemap index, for example https://example.com/sitemap.xml.

Output parameters:

- `result` (string)

### `explain_finding` (~121 tokens)

Explain a finding id (e.g. access.blocked, robots.disallowed, schema.software_no_offers) in the given language: what it means, why it matters for AI search and how to fix it.

Input parameters:

- `finding_id` (string, required): Finding id exactly as shown in an audit report, for example access.blocked or llms.missing.
- `language`: Report language as a code such as en, de, fr, es, ja or zh. Defaults to the account language (English without a key); 60 languages are supported.

Output parameters:

- `result` (string)

### `compare_urls` (~123 tokens)

Audit 2 or 3 URLs (yours and competitors') and compare them side by side: score, P1/P2/P3 counts, blocked crawlers, llms.txt, structured data, text without JavaScript. Needs an API key.

Input parameters:

- `language`: Report language as a code such as en, de, fr, es, ja or zh. Defaults to the account language (English without a key); 60 languages are supported.
- `urls` (array, required): 2 or 3 page URLs or domains to compare; put your own page first.

Output parameters:

- `result` (string)

### `crawl_site` (~193 tokens)

SEO site audit: crawl up to 50 pages (up to 90 seconds) of a site from a start page by following internal links (then the sitemap), as a browser and respecting robots.txt. Reports broken and redirected internal links, duplicate titles and descriptions, orphan, deep and thin pages, sitemap entries that are noindex, redirect or name another canonical, and on-page SEO problems counted across pages. Needs an API key.

Input parameters:

- `language`: Report language as a code such as en, de, fr, es, ja or zh. Defaults to the account language (English without a key); 60 languages are supported.
- `max_pages` (integer): How many pages to crawl, 1 to 50.
- `url` (string, required): Start page of the crawl, usually the home page: a full URL or a bare domain such as example.com. Only pages on the same host are crawled.

Output parameters:

- `result` (string)

### `find_broken_links` (~195 tokens)

Broken link finder: read a page and the pages of the same site it links to, level by level (depth 0 is the start page only; up to 50 pages and 90 seconds), and check every link found once: pages, other sites and images. Returns the broken links with the link text and the pages that contain them, apart from links that could not be verified (blocked, rate limited, slow) and redirects. Respects robots.txt for pages it walks into. Needs an API key.

Input parameters:

- `depth` (integer): Levels of pages to follow from the start page, 0 to 3.
- `max_pages` (integer): How many pages to read, 1 to 50.
- `url` (string, required): Start page: a full URL or a bare domain such as example.com. Only pages on the same host are walked; links to other sites are checked but not followed.

Output parameters:

- `result` (string)

### `get_all_pages_performance` (~198 tokens)

Every page of the site with clicks, impressions, CTR, position, trend, page category (winner, zero click, high impressions low CTR, declining, cannibalized, not indexable) and recommended action. Uses your connected Google Search Console (read-only) plus a crawl of the site; the analysis is reused for a day. Never changes your site: redirects are proposals only. Needs an API key from a paid plan.

Input parameters:

- `category`: Only pages in this category: winner, zero_click, high_impressions_low_ctr, declining, cannibalized, low_traffic, not_indexable, no_data.
- `limit` (integer): Rows to return, 1 to 200.
- `site`: Search Console property, for example sc-domain:example.com or https://example.com/. Omit to use the site of your latest Page Intelligence analysis.
- `sort` (string): clicks, impressions or position.

Output parameters:

- `result` (string)

### `find_zero_click_pages` (~168 tokens)

Pages with zero recorded clicks in the last 90 days, each with a verdict (KEEP, OPTIMIZE, MERGE, REDIRECT or REVIEW FOR REMOVAL) and the evidence. A trigger for review, never for automatic deletion; pages Search Console returned nothing for are reported as unknown, not as zero. Uses your connected Google Search Console (read-only) plus a crawl of the site; the analysis is reused for a day. Never changes your site: redirects are proposals only. Needs an API key from a paid plan.

Input parameters:

- `limit` (integer): Rows to return, 1 to 200.
- `site`: Search Console property, for example sc-domain:example.com or https://example.com/. Omit to use the site of your latest Page Intelligence analysis.

Output parameters:

- `result` (string)

### `find_high_impression_low_ctr_pages` (~142 tokens)

Pages Google shows often that users rarely click, with the query, the position, the CTR that is normal there and what is wrong with the title, description or query match. Uses your connected Google Search Console (read-only) plus a crawl of the site; the analysis is reused for a day. Never changes your site: redirects are proposals only. Needs an API key from a paid plan.

Input parameters:

- `limit` (integer): Rows to return, 1 to 200.
- `site`: Search Console property, for example sc-domain:example.com or https://example.com/. Omit to use the site of your latest Page Intelligence analysis.

Output parameters:

- `result` (string)

### `find_declining_pages` (~142 tokens)

Pages losing organic clicks against the previous comparable period, with the likely cause (ranking loss, lower demand, cannibalization, CTR loss, technical indexing, outdated content) and the action. Uses your connected Google Search Console (read-only) plus a crawl of the site; the analysis is reused for a day. Never changes your site: redirects are proposals only. Needs an API key from a paid plan.

Input parameters:

- `limit` (integer): Rows to return, 1 to 200.
- `site`: Search Console property, for example sc-domain:example.com or https://example.com/. Omit to use the site of your latest Page Intelligence analysis.

Output parameters:

- `result` (string)

### `detect_keyword_cannibalization` (~147 tokens)

Groups of URLs competing for the same search queries: shared queries, intent and content similarity, the primary page, the recommended action and a confidence score. Separates harmful cannibalization from legitimate multiple rankings. Uses your connected Google Search Console (read-only) plus a crawl of the site; the analysis is reused for a day. Never changes your site: redirects are proposals only. Needs an API key from a paid plan.

Input parameters:

- `limit` (integer): Rows to return, 1 to 200.
- `site`: Search Console property, for example sc-domain:example.com or https://example.com/. Omit to use the site of your latest Page Intelligence analysis.

Output parameters:

- `result` (string)

### `identify_strongest_page` (~169 tokens)

The strongest URL of each group of competing pages, scored on clicks, history, query coverage, ranking stability, intent match, content, internal links, canonical setup and business value, with every component shown (the page with the most clicks does not always win). Uses your connected Google Search Console (read-only) plus a crawl of the site; the analysis is reused for a day. Never changes your site: redirects are proposals only. Needs an API key from a paid plan.

Input parameters:

- `limit` (integer): Rows to return, 1 to 200.
- `site`: Search Console property, for example sc-domain:example.com or https://example.com/. Omit to use the site of your latest Page Intelligence analysis.
- `url`: Only the group that contains this URL.

Output parameters:

- `result` (string)

### `find_content_consolidation_candidates` (~139 tokens)

Groups of pages that should be merged or redirected into a stronger page, with what to do with each competing page and why. Pages with a distinct intent are kept. Uses your connected Google Search Console (read-only) plus a crawl of the site; the analysis is reused for a day. Never changes your site: redirects are proposals only. Needs an API key from a paid plan.

Input parameters:

- `limit` (integer): Rows to return, 1 to 200.
- `site`: Search Console property, for example sc-domain:example.com or https://example.com/. Omit to use the site of your latest Page Intelligence analysis.

Output parameters:

- `result` (string)

### `generate_redirect_plan` (~164 tokens)

A 301 redirect plan for redundant pages: source, destination, reason, Search Console evidence, intent similarity, confidence, risk, required content migration and internal links to update. Detects chains, loops, canonical conflicts and irrelevant destinations. Proposals only: nothing is executed and every redirect needs your approval. Uses your connected Google Search Console (read-only) plus a crawl of the site; the analysis is reused for a day. Never changes your site: redirects are proposals only. Needs an API key from a paid plan.

Input parameters:

- `limit` (integer): Rows to return, 1 to 200.
- `site`: Search Console property, for example sc-domain:example.com or https://example.com/. Omit to use the site of your latest Page Intelligence analysis.

Output parameters:

- `result` (string)

### `generate_content_merge_plan` (~139 tokens)

For each consolidation group: the content and queries to move into the target page, improvements to the target, internal links to update, sitemap and canonical changes and a validation plan. Uses your connected Google Search Console (read-only) plus a crawl of the site; the analysis is reused for a day. Never changes your site: redirects are proposals only. Needs an API key from a paid plan.

Input parameters:

- `limit` (integer): Rows to return, 1 to 200.
- `site`: Search Console property, for example sc-domain:example.com or https://example.com/. Omit to use the site of your latest Page Intelligence analysis.

Output parameters:

- `result` (string)

### `prioritize_seo_opportunities` (~155 tokens)

A ranked P1-P3 action queue to grow organic traffic from existing pages (consolidate, fix snippets, recover declines, fix indexing, add internal links) with a transparent score, exact steps and impact split into measured numbers and labelled estimates. Uses your connected Google Search Console (read-only) plus a crawl of the site; the analysis is reused for a day. Never changes your site: redirects are proposals only. Needs an API key from a paid plan.

Input parameters:

- `limit` (integer): Rows to return, 1 to 200.
- `site`: Search Console property, for example sc-domain:example.com or https://example.com/. Omit to use the site of your latest Page Intelligence analysis.

Output parameters:

- `result` (string)

### `watch_site` (~104 tokens)

Add a URL to the account's scheduled monitoring (freq: daily or weekly). The owner gets an email when a check finds a new P1 problem or the score drops by 10 or more. Needs an API key; adding the same URL again only changes the frequency.

Input parameters:

- `freq` (string): How often to check the page: daily or weekly (default).
- `url` (string, required): Page to monitor: a full URL or a bare domain such as example.com.

Output parameters:

- `result` (string)

### `list_watched_sites` (~33 tokens)

List the URLs this account monitors, with the frequency, last score and last check time. Needs an API key.

Output parameters:

- `result` (string)

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/com-aiseewhy-aiseewhy/api#diagnostics

## Score history

- 2026-11-21: 77
- 2026-11-20: 76
- 2026-11-19: 75
- 2026-11-18: 75
- 2026-11-17: 75
- 2026-11-16: 74
- 2026-11-15: 74
- 2026-11-14: 74
- 2026-11-13: 73
- 2026-11-12: 73

## Common questions

### What is the AISeeWhy MCP server?

AISeeWhy is an MCP server listed in the public MCP registry as com.aiseewhy/aiseewhy. Can ChatGPT, Claude and Perplexity crawlers read your page? Audit, fixes, compare, monitoring. This page covers its hosted endpoint (https://api.aiseewhy.com/mcp).

### Is the AISeeWhy MCP server safe to use?

AISeeWhy scores 77 out of 100 on VerifyMCP. 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 AISeeWhy MCP server expose?

AISeeWhy exposes 19 tools: show_ai_crawler_check, audit_url, audit_sitemap, explain_finding, compare_urls, and 14 more. Their descriptions and schemas cost roughly 2,849 tokens of context every time the server is loaded.

### Does the AISeeWhy MCP server require authentication?

No. We connected to AISeeWhy without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the AISeeWhy MCP server still maintained?

AISeeWhy is still listed as active in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What did VerifyMCP test on the AISeeWhy MCP server?

VerifyMCP ran 14 checks on AISeeWhy, covering TLS certificate, Authorization, HTTPS, HSTS header and 10 more. Supply Chain Security and Provenance & Transparency do not apply to a hosted endpoint, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.

## Links

- Remote endpoint: https://api.aiseewhy.com/mcp
- Repository: https://github.com/aiseewhy/check-action
- Website: https://aiseewhy.com/docs/#mcp
- Changelog RSS feed: https://verifymcp.io/servers/com-aiseewhy-aiseewhy/api.xml
- Changelog JSON feed: https://verifymcp.io/servers/com-aiseewhy-aiseewhy/api/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/com-aiseewhy-aiseewhy/api.json
- HTML version of this page: https://verifymcp.io/servers/com-aiseewhy-aiseewhy/api
