# gl.parse/mcp (remote · agent.parse.gl)

How ChatGPT and Google AI Mode name, recommend and cite brands.

- Trust score: 74/100 (medium)
- Change this week: +5
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-21

## Components

- remote · `agent.parse.gl`: 74/100 (this document), [markdown](https://verifymcp.io/servers/gl-parse-mcp/agent.md), [page](https://verifymcp.io/servers/gl-parse-mcp/agent)
- npm · `@parse-gl/mcp`: 33/100, [markdown](https://verifymcp.io/servers/gl-parse-mcp/parse-gl-mcp.md), [page](https://verifymcp.io/servers/gl-parse-mcp/parse-gl-mcp)

## Channel facts

- Endpoint: `https://agent.parse.gl/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.1.0`

## 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**: 69/100
  - 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 2382 tokens (~397/item across 6 items; 6 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 12/100
  - Stability check failed: schema churn in the 8 days we've observed: 1 tool removals, 0 breaking changes, 0 auth/transport breaks, 0 additions.
- **Tool Coverage**: 98/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 94% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 6 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 7 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the gl.parse/mcp server?

gl.parse/mcp is a hosted endpoint at https://agent.parse.gl/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 gl-parse-mcp 'https://agent.parse.gl/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "gl-parse-mcp": {
      "url": "https://agent.parse.gl/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "gl-parse-mcp": {
      "type": "http",
      "url": "https://agent.parse.gl/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.gl-parse-mcp]
url = "https://agent.parse.gl/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "gl-parse-mcp": {
      "type": "remote",
      "url": "https://agent.parse.gl/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add gl-parse-mcp --url 'https://agent.parse.gl/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  gl-parse-mcp:
    url: "https://agent.parse.gl/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "gl-parse-mcp": {
      "Transport": "http",
      "Url": "https://agent.parse.gl/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add gl-parse-mcp -t streamable-http -u 'https://agent.parse.gl/mcp'
```

### Other

```json
{
  "mcpServers": {
    "gl-parse-mcp": {
      "type": "http",
      "url": "https://agent.parse.gl/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 74, +4)

- [security] Tool “search_cited_pages” rewrote its description, which is the text the model reads
- [security] Tool “query_metrics” rewrote its description, which is the text the model reads
- [functional improvement] Schema quality: unverified → 100
- [functional] The server now declares the “prompts” capability
- [functional] The server now declares the “resources” capability
- [functional] New prompt “start”
- [functional] Server version: daed0a57e09f → bef7439da312
- [cosmetic] “search_cited_pages” reworded the description of “pages”
- [cosmetic] “get_brief” reworded the description of “kinds”

### 2026-11-20 (score 70, +1)

- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “search_answers” rewrote its description, which is the text the model reads
- [security] Tool “search_cited_pages” rewrote its description, which is the text the model reads
- [security] Tool “check_brand” rewrote its description, which is the text the model reads
- [security] Tool “query_metrics” rewrote its description, which is the text the model reads
- [security] Tool “resolve_ids” rewrote its description, which is the text the model reads
- [functional improvement] Schema quality: 446 → 392
- [functional] Server version: f1a4a54a2581 → daed0a57e09f
- [cosmetic] “search_answers” added an optional parameter “history”
- [cosmetic] “search_cited_pages” added an optional parameter “history”
- [cosmetic] “query_metrics” added an optional parameter “history”
- [cosmetic] “resolve_ids” reworded the description of “q”
- [cosmetic] “get_brief” reworded the description of “kinds”

### 2026-11-19 (score 69, 0)

- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “check_brand” rewrote its description, which is the text the model reads
- [security] Tool “query_metrics” rewrote its description, which is the text the model reads
- [security] Tool “resolve_ids” rewrote its description, which is the text the model reads
- [security] Tool “search_answers” rewrote its description, which is the text the model reads
- [security] Tool “search_cited_pages” rewrote its description, which is the text the model reads
- [functional improvement] Schema quality: 503 → 446
- [functional] Server version: e5bfaeba3303 → f1a4a54a2581
- [cosmetic] “resolve_ids” reworded the description of “q”
- [cosmetic] “query_metrics” reworded the description of “explain”
- [cosmetic] “resolve_ids” reworded the description of “kinds”
- [cosmetic] “query_metrics” dropped the optional parameter “fields”
- [cosmetic] “query_metrics” dropped the optional parameter “format”
- [cosmetic] “query_metrics” dropped the optional parameter “level”
- [cosmetic] “query_metrics” dropped the optional parameter “nudges”
- [cosmetic] “query_metrics” dropped the optional parameter “panel”
- [cosmetic] “search_answers” dropped the optional parameter “explain”
- [cosmetic] “search_answers” dropped the optional parameter “format”
- [cosmetic] “search_answers” dropped the optional parameter “level”
- [cosmetic] “search_answers” dropped the optional parameter “nudges”
- [cosmetic] “search_answers” dropped the optional parameter “view”
- [cosmetic] “search_cited_pages” dropped the optional parameter “explain”
- [cosmetic] “search_cited_pages” dropped the optional parameter “format”
- [cosmetic] “search_cited_pages” dropped the optional parameter “level”
- [cosmetic] “search_cited_pages” dropped the optional parameter “minRetrievedBy”
- [cosmetic] “search_cited_pages” dropped the optional parameter “nudges”

### 2026-11-18 (score 69, 0)

- [functional] Server version: 35d86653bede → e5bfaeba3303
- [cosmetic] “search_cited_pages” reworded the description of “listsBrand”
- [cosmetic] “get_brief” reworded the description of “kinds”

### 2026-11-17 (score 69, 0)

- [security] The server rewrote its instructions, which are the text every model session reads
- [functional] Server version: 40f2583e047d → 35d86653bede

### 2026-11-16 (score 69, 0)

- [functional] Server version: 5e933e702285 → 40f2583e047d

### 2026-11-15 (score 69, 0)

- [security regression] Stability: 0.03 → fail
- [security regression] Tool “list_facets” was removed
- [security] Tool “query_metrics” rewrote its description, which is the text the model reads
- [security] Tool “resolve_ids” rewrote its description, which is the text the model reads
- [security] Tool “search_answers” rewrote its description, which is the text the model reads
- [security] Tool “check_brand” rewrote its description, which is the text the model reads
- [functional regression] Schema quality: 408 → 493
- [functional improvement] Tool coverage: 83% → 90%
- [functional] Server version: 6d436f8d3e22 → 5e933e702285
- [cosmetic] “get_brief” added an optional parameter “cursor”
- [cosmetic] “get_brief” added an optional parameter “kinds”
- [cosmetic] “get_brief” added an optional parameter “limit”
- [cosmetic] “get_brief” added an optional parameter “until”
- [cosmetic] “get_brief” added an optional parameter “view”
- [cosmetic] “resolve_ids” added an optional parameter “match”
- [cosmetic] “resolve_ids” added an optional parameter “unit”
- [cosmetic] “search_answers” added an optional parameter “rows”
- [cosmetic] “search_cited_pages” added an optional parameter “view”
- [cosmetic] “get_brief” reworded the description of “since”
- [cosmetic] “query_metrics” reworded the description of “measures”
- [cosmetic] “resolve_ids” reworded the description of “kinds”
- [cosmetic] “resolve_ids” reworded the description of “q”
- [cosmetic] “search_answers” reworded the description of “sort”
- [cosmetic] “search_answers” reworded the description of “view”
- [cosmetic] “query_metrics” made “measures” optional
- [cosmetic] “resolve_ids” made “q” optional
- [cosmetic] Tool “resolve_ids” changed its title: Resolve names, domains and URLs to ids → Resolve names to ids, find by meaning, list facets
- [cosmetic] Tool “search_answers” changed its title: Search AI answers → Search AI answers, or what they say
- [cosmetic] Tool “search_cited_pages” changed its title: Search pages AI answers cite → Search pages AI cites, and search result lists
- [cosmetic] Tool “get_brief” changed its title: Get the brief → Get the brief, or the event log

### 2026-11-14 (score 69, 0)

- [security] Tool “check_brand” rewrote its description, which is the text the model reads
- [security] Tool “query_metrics” rewrote its description, which is the text the model reads
- [security] Tool “search_answers” rewrote its description, which is the text the model reads
- [security] Tool “search_cited_pages” rewrote its description, which is the text the model reads
- [functional improvement] Stability: unverified → 0.03
- [functional] Server version: c7dafc3e2292 → 6d436f8d3e22
- [cosmetic] “get_brief” reworded the description of “moves”
- [cosmetic] “query_metrics” reworded the description of “window”
- [cosmetic] “search_answers” reworded the description of “window”
- [cosmetic] “search_cited_pages” reworded the description of “pages”
- [cosmetic] “search_cited_pages” reworded the description of “window”
- [cosmetic] “search_cited_pages” dropped the optional parameter “brands”

## MCP tools (6)

### `check_brand` (~163 tokens)

Check a brand: who it is, where it stands, what comes next

Look up how AI answers talk about a brand, who it loses to, verified product facts, and AI crawler blocks. Call this first for brand positioning, competitor research, AI visibility, or before writing copy about a brand. Takes a website (best), brand name, or brd_ id. Returns data.kind (measured, thin, unplaced, which, unseen): for measured brands, Recommended share (with share_named detail) per market, top rivals, and weakest topic; for ambiguous names (which), the candidate list without reading; for thin/unseen, coverage status and next steps. No brand score. Free.

Input parameters:

- `subject` (string, required): the brand's website (best: it settles which company), its name, or a brd_ id

### `resolve_ids` (~225 tokens)

Resolve names to ids, find by meaning, list facets

Resolve up to 5 names, domains, URLs, or question wordings to Parse ids, with match scores and brand index status. Use kinds: ["facet"], ["surface"], or ["measure"] for filter values, AI surfaces, or metric definitions. For tool arguments pass kinds: ["argument"] and tool names in q. Free.

Input parameters:

- `kinds` (array): what to match: brand, topic, question, market, page, site, place (default all); or facet, surface, measure alone for the dictionary; argument alone, with tool names in q: what a tool takes beyond its…
- `match` (string): name: exact names, domains and ids (the search by meaning needs sign-in)
- `q` (array): up to 5 names, domains, URLs, question wordings or typed ids; the dictionary takes search words or none; kinds argument takes tool names
- `unit` (string): dictionary only: facets of this unit (answer, mention, evidence, comparison, page, page_brand, question, topic_brand)

### `query_metrics` (~473 tokens)

Query metrics over AI answers

Measure which brands AI names or recommends across topics, markets, or engines. Public measures: answers, questions, share_named, share_recommended, listing_share, concentration (concentration takes no by). Rows include answer and question counts. Shares include a 95% range when available. For cited pages use search_cited_pages. Free. More arguments: resolve_ids {"kinds":["argument"],"q":["query_metrics"]}.

Input parameters:

- `brands` (array): with by brand: only these brands' rows, against each other in the same answers (in place of the most named)
- `by` (array): up to 3 of brand, topic, market, question, engine, surface_class, place, country, day, week, month, site, page, page_type, axis, versus_brand, or facet:<key>
- `compare` (string): prior: the window of equal length before, with significance; none: no comparison. Write the value bare, compare: prior, with no quotes inside the string
- `cursor` (string)
- `explain`: false: leave out the one-sentence note (data.note). change (with subject, a share and compare prior): why each share moved, split into the same questions and the question mix, with engine model chang…
- `filter` (object): which answers count; window is a separate argument, not part of filter
- `history` (string): answers past 90 days, $0.001 each once: auto (default) a sample worth up to a quarter of the month's budget, at most $5; all, every answer within the budget; free, none, with each sample's cost
- `limit` (integer): default 25; the token cap wins
- `measures` (array): required: what to count
- `sort` (string): a measure or a by dimension; "<name> asc" or "<name> desc"
- `subject` (string): brd_ id the shares are about; or group by brand. With no filter it reads the brand's own markets (up to three) pooled, named in context.marketsRead: pass filter.market for one market's number
- `window` (object): default the last 28 days; "last" or from/to

### `search_answers` (~302 tokens)

Search AI answers, or what they say

Find AI answers behind a visibility number. Filter by brand, topic, market, question, engine, place, country, citations, keywords, or other facets. Rows include ans_ ids. Use view: "diff" to see brand and citation changes for up to five questions or one topic. For counts and shares use query_metrics. Free. More arguments: resolve_ids {"kinds":["argument"],"q":["search_answers"]}.

Input parameters:

- `brands` (array): brd_ ids to mark in text
- `cursor` (string)
- `fields` (array): columns to keep in each row
- `filter` (object): which answers count; window is a separate argument, not part of filter
- `history` (string): view: "diff" only: answers past 90 days, $0.001 each once: auto (default) a sample worth up to a quarter of the month's budget, at most $5; all, every answer within the budget; free, none, with each…
- `limit` (integer): default 25; the token cap wins
- `rows` (string): answers: one row an answer (the sentences and comparisons inside them need sign-in)
- `sort` (string): answers only: relevance (with filter.text; default then), date (default), position (with filter.brand)
- `window` (object): default the last 28 days; "last" or from/to

### `search_cited_pages` (~415 tokens)

Search pages AI cites, and search result lists

Find web pages AI cites by brand, topic, market, question, engine, site, or facet. For outreach pass pages: "third_party", filter.brand, and listsBrand: false for pages missing the brand. filter.brandRule: "not_named" selects the answers that omit it. Rows show citation counts, page kind, and whether the page lists that brand when available. Free. More arguments: resolve_ids {"kinds":["argument"],"q":["search_cited_pages"]}.

Input parameters:

- `cursor` (string)
- `fields` (array): columns to keep in each row
- `filter` (object): which answers count; window is a separate argument, not part of filter
- `history` (string): answers past 90 days, $0.001 each once: auto (default) a sample worth up to a quarter of the month's budget, at most $5; all, every answer within the budget; free, none, with each sample's cost
- `limit` (integer): default 25; the token cap wins
- `listsBrand` (boolean): needs filter.brand: true keeps pages the page judge says list the brand; false, pages that do not: by the judge, or a stored copy that never names it. With brandRule not_named, false is the pages to…
- `pages` (string): all, or third_party: only pages on no brand's own site (not the filter brand's, not any other brand's) and not Parse's own. For a listing pitch, also pass filter.brand and listsBrand: false. Write th…
- `sort` (string): default citations; omitting needs filter.brand
- `view` (string): pages: pages AI answers cite (search APIs' ranked lists need sign-in)
- `window` (object): default the last 28 days; "last" or from/to

### `get_brief` (~462 tokens)

Get the brief, or the event log

Needs a free Parse sign-in, which this call asks the person for: the person's own brand's brief (what changed, what to do next) and their work record. Call it only when the person wants their own brand, tracking or their record, or says yes to an offer; for any public question use check_brand and the other tools.

Input parameters:

- `brand` (string): the brd_ id of the brand this job is for, in a workspace that serves several (an agency's clients): the brief is about it, nobody is asked which brand is theirs, and nothing is claimed or billed
- `chains` (boolean): add the common jobs, each as its chain of calls (always added while the workspace has no brand)
- `cursor` (string): events only: the page.cursor (an event id) to continue after
- `kinds` (array): events only: only these kinds: mention_move, rival_move, finding, fact_contradicted, page_first_cited, page_dropped, page_changed, proof_step_blocked, own_page_changed, action_outcome, new_answer_bat…
- `limit` (integer): events only: default 25
- `moves` (boolean): add every brand of the workspace (its own and those it watches), each in its own markets, this week against the week before, largest move first, one read for all
- `peek` (boolean): read the brief without marking it read, so the next call (this key's or another's) still starts where the last read did
- `sections` (array): default all
- `since`: "last" (default: since the last brief) or an ISO date or time; with view: "events", a day (UTC), in place of the last 7 days
- `until` (string): events only: only events on or before this day (UTC)
- `view` (string): brief (default): what passed the bar, ranked; events: every event

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/gl-parse-mcp/agent#diagnostics

## Score history

- 2026-11-21: 74
- 2026-11-20: 70
- 2026-11-19: 69
- 2026-11-18: 69
- 2026-11-17: 69
- 2026-11-16: 69
- 2026-11-15: 69
- 2026-11-14: 69
- 2026-11-13: 69
- 2026-11-12: 39

## Common questions

### What is the gl.parse/mcp server?

gl.parse/mcp is listed in the public MCP registry as gl.parse/mcp. How ChatGPT and Google AI Mode name, recommend and cite brands. This page covers its hosted endpoint (https://agent.parse.gl/mcp).

### Is the gl.parse/mcp server safe to use?

gl.parse/mcp scores 74 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 gl.parse/mcp server expose?

gl.parse/mcp exposes 6 tools: check_brand, resolve_ids, query_metrics, search_answers, search_cited_pages, get_brief. Their descriptions and schemas cost roughly 2,040 tokens of context every time the server is loaded.

### Does the gl.parse/mcp server require authentication?

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

### Is the gl.parse/mcp server still maintained?

gl.parse/mcp 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 gl.parse/mcp server?

VerifyMCP ran 13 checks on gl.parse/mcp, covering TLS certificate, Authorization, HTTPS, HSTS header and 9 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://agent.parse.gl/mcp
- Website: https://parse.gl/docs/mcp
- Changelog RSS feed: https://verifymcp.io/servers/gl-parse-mcp/agent.xml
- Changelog JSON feed: https://verifymcp.io/servers/gl-parse-mcp/agent/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/gl-parse-mcp/agent.json
- HTML version of this page: https://verifymcp.io/servers/gl-parse-mcp/agent
