browser-compat-mcp-server

Browsercompat Search Features

browsercompat_search_features

Read-onlyIdempotent

Find web features by plain name or keyword when the canonical key is unknown, across CSS, JavaScript, HTML, Web APIs, SVG, MathML, WebAssembly, and HTTP headers. Returns ranked matches with the BCD key, the web-features id, the Baseline state, a one-line support summary, and which field matched. Feed a result key into browsercompat_get_feature for the full record.

Input Schema

Name	Required	Description
`group`	No	Restrict results to features in one web-features group or any group nested under it, for example "selectors" or "css". Keys with no web-features mapping never match. Call browsercompat_list_reference with topic groups for valid ids.
`limit`	No	Maximum results to return, at most 50 per call; a value from 51 to 1000 is served as 50. When matches remain, cap reports the size applied and nextOffset continues.
`query`	Yes	Plain-language name, keyword, or code notation, for example "container query", "fromAsync", "Array.prototype.at", "display: grid", or "<dialog>".
`offset`	No	Number of ranked matches to skip before returning results, for paging past limit. Pass the nextOffset from the previous response.
`baseline`	No	Restrict results to one Baseline state.
`snapshot`	No	Restrict results to features in one ECMAScript snapshot, for example "ecmascript-2023". Call browsercompat_list_reference with topic snapshots for valid ids.
`namespace`	No	Restrict results to one top-level browser-compat-data namespace.

Output Schema

Name	Required	Description
`cap`	No	The page size applied: limit, or 50 when limit was above 50.
`error`	No	Present when the call failed. Absent on success.
`shown`	No	Number of results on this page.
`results`	No	Ranked matches, best first. An empty array is a successful search with no hits.
`truncated`	No	True when more matches remain past this page.
`nextOffset`	No	The offset of the next page. Present only while matches remain past this page.
`totalCount`	No	Matches across every page, before offset and limit apply.
`data_version`	No	Vintage of each bundled dataset behind this answer.
`offsetNotice`	No	Why a page came back empty although the search matched.
`noMatchNotice`	No	How to broaden a search that matched nothing.
`appliedFilters`	No	Filters the server applied to this search.

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedOct 26, 2026

changedOutput schema / properties / results / items / properties / support_summary / description

Previous value: -"One line over the Baseline core browsers, with — for unsupported and ? for unknown."New value: +"One line over the Baseline core browsers: addition release with ≤ for an upper bound and (partial), (prefixed), or (flagged) for qualified support; — for unsupported, removed, or preview-only; ? for unknown."

Changed3 schema fields changedOct 15, 2026

changedInput schema / properties / limit / description

Previous value: -"Maximum results to return, from 1 to 50."New value: +"Maximum results to return, at most 50 per call; a value from 51 to 1000 is served as 50. When matches remain, cap reports the size applied and nextOffset continues."

changedInput schema / properties / limit / maximum
```
Previous value: -50New value: +1000
```

changedOutput schema / properties / cap / description

Previous value: -"The limit that was applied."New value: +"The page size applied: limit, or 50 when limit was above 50."

Changed15 schema fields changedSep 22, 2026

addedInput schema / properties / group

Added value: +{
+  "description": "Restrict results to features in one web-features group or any group nested under it, for example \"selectors\" or \"css\". Keys with no web-features mapping never match. Call browsercompat_list_reference with topic groups for valid ids.",
+  "maxLength": 100,
+  "minLength": 1,
+  "type": "string"
+}

addedInput schema / properties / offset

Added value: +{
+  "default": 0,
+  "description": "Number of ranked matches to skip before returning results, for paging past limit. Pass the nextOffset from the previous response.",
+  "maximum": 9007199254740991,
+  "minimum": 0,
+  "type": "integer"
+}

changedInput schema / properties / query / description

Previous value: -"Plain-language name or keyword, for example \"container query\" or \"fromAsync\"."New value: +"Plain-language name, keyword, or code notation, for example \"container query\", \"fromAsync\", \"Array.prototype.at\", \"display: grid\", or \"<dialog>\"."

addedInput schema / properties / snapshot

Added value: +{
+  "description": "Restrict results to features in one ECMAScript snapshot, for example \"ecmascript-2023\". Call browsercompat_list_reference with topic snapshots for valid ids.",
+  "maxLength": 100,
+  "minLength": 1,
+  "type": "string"
+}

addedOutput schema / properties / appliedFilters / properties / group

Added value: +{
+  "description": "web-features group filter the search applied, nested groups included.",
+  "type": "string"
+}

addedOutput schema / properties / appliedFilters / properties / snapshot

Added value: +{
+  "description": "ECMAScript snapshot filter the search applied.",
+  "type": "string"
+}

changedOutput schema / properties / error / properties / data / properties / reason / description

Previous value: -"Machine-readable failure mode. Declared by this tool: `invalid_query`: query is whitespace-only after normalization, or normalizes to zero tokens. Other values are possible when a failure originates below the handler."New value: +"Machine-readable failure mode. Declared by this tool: `invalid_query`: query is whitespace-only after normalization, or normalizes to zero tokens. `unknown_group`: group names no web-features group in the bundled data. `unknown_snapshot`: snapshot names no ECMAScript snapshot in the bundled data. Other values are possible when a failure originates below the handler."

changedOutput schema / properties / error / properties / data / properties / reason / examples

Previous value: -[
-  "invalid_query"
-]New value: +[
+  "invalid_query",
+  "unknown_group",
+  "unknown_snapshot"
+]

addedOutput schema / properties / nextOffset

Added value: +{
+  "description": "The offset of the next page. Present only while matches remain past this page.",
+  "type": "number"
+}

addedOutput schema / properties / offsetNotice

Added value: +{
+  "description": "Why a page came back empty although the search matched.",
+  "type": "string"
+}

changedOutput schema / properties / results / items / properties / matched_on / description

Previous value: -"Which field matched the query, so the ranking can be inspected."New value: +"Which field matched the query, so the ranking can be inspected. path_suffix means the key ends in the dotted or property-value notation the query used, such as Element.animate or display: grid."

changedOutput schema / properties / results / items / properties / matched_on / enum

Previous value: -[
-  "bcd_key",
-  "baseline_id",
-  "name",
-  "caniuse_title",
-  "path_segment",
-  "description",
-  "path_tokens"
-]New value: +[
+  "bcd_key",
+  "baseline_id",
+  "name",
+  "caniuse_title",
+  "path_suffix",
+  "path_segment",
+  "description",
+  "path_tokens"
+]

changedOutput schema / properties / shown / description

Previous value: -"Number of results returned."New value: +"Number of results on this page."

changedOutput schema / properties / totalCount / description

Previous value: -"Matches before the display cap was applied."New value: +"Matches across every page, before offset and limit apply."

changedOutput schema / properties / truncated / description

Previous value: -"True when matches exceeded limit."New value: +"True when more matches remain past this page."

First observedSep 20, 2026

TDQS

A4.2/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnly/idempotent/closed-world, so safety is covered. The description adds real context beyond that: what each result carries (BCD key, web-features id, Baseline state, support summary, matched field) and that results are ranked, which tells the agent how to interpret ordering. Pagination behavior is left to the schema rather than the description.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Three sentences, zero filler: purpose and scope first, return shape second, workflow hand-off last. Every sentence earns its place and nothing is buried.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With an output schema present, return values need not be spelled out, yet the description summarizes them helpfully and closes the loop to get_feature. The remaining gap is silent about the filtering capability exposed by four of the seven parameters, which an agent would only discover by reading the schema.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so baseline is 3. The description adds query-shape intuition ("plain name or keyword", canonical key unknown) but never mentions the filter parameters (group, baseline, snapshot, namespace) that materially narrow results, so it contributes little beyond the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ("Find web features by plain name or keyword") and names the exact scope: when the canonical key is unknown, across the enumerated standards areas. It is clearly distinguishable from browsercompat_get_feature, which the description names as the follow-up for a known key.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives an explicit trigger ("when the canonical key is unknown") and an explicit next step ("Feed a result key into browsercompat_get_feature for the full record"), which routes the agent between siblings. It stops short of stating the inverse exclusion — that a known key should go straight to get_feature — leaving that to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All browser-compat-mcp-server tools Inspect

Try in Browser

Shareable, User-Owned Context BackendPeople, applications, and agents read, write, and coordinate together in one context lake. Reach it over MCP, a CLI, a Python SDK, or REST.fulcradynamics.comAd

Glama MCP Gateway

Add one secure layer between your agents and this server.