Firmware FLASH/RAM history, build diffs, symbol sizes and benchmarks for embedded projects
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
MemBrowse account API key, for private projects. Omit it to read public projects by their GitHub owner/repo.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Compare two builds of a target. Always returns each region's old and new size; 'level' adds the changed sections, source files or symbols, sorted by size of change. Without 'base', the head is compared with the build it was compared with when it was uploaded: for a pull request's head commit that is the tip of the PR's base branch, so the result is the whole PR; for a pushed commit it is the branch's previous build, so the result is what that one commit did. With 'base_target', the base is a build of another target instead, to see which target already does without a symbol.
Target name (see list_targets).
The project: a public project's GitHub 'owner/repo' (no API key needed), or, with an API key, a project id or name from list_projects.
The older build, in the same forms. Default: the build 'head' was compared with when it was uploaded (for a pull request, the tip of its base branch); with 'base_target', the head of that target's default branch.
The newer build: a full commit SHA, a SHA prefix (7+ characters), or a branch or tag name. Default: the head of the default branch. For a pull request, its head commit SHA.
Detail to return (default 'regions'). 'symbols' and 'files' load full reports and are slower.
Maximum rows to return (default 50, max 200).
Only this memory region, e.g. FLASH or RAM.
Only this section, e.g. .text (sections and symbols levels).
Compare with this other target of the project ('base' then names its build).
Only sections, files or symbols whose name contains this text (case-insensitive).
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http membrowse https://api.membrowse.com/mcp
Headers
Authorization (optional, not in the config above): MemBrowse account API key, for private projects. Omit it to read public projects by their GitHub owner/repo.Compare two builds of a target. Always returns each region's old and new size; 'level' adds the changed sections, source files or symbols, sorted by size of change. Without 'base', the head is compared with the build it was compared with when it was uploaded: for a pull request's head commit that is the tip of the PR's base branch, so the result is the whole PR; for a pushed commit it is the branch's previous build, so the result is what that one commit did. With 'base_target', the base is a build of another target instead, to see which target already does without a symbol.
| Argument | Type | Description |
|---|---|---|
| base | string | The older build, in the same forms. Default: the build 'head' was compared with when it was uploaded (for a pull request, the tip of its base branch); with 'base_target', the head of that target's default branch. |
| head | string | The newer build: a full commit SHA, a SHA prefix (7+ characters), or a branch or tag name. Default: the head of the default branch. For a pull request, its head commit SHA. |
| level | string | Detail to return (default 'regions'). 'symbols' and 'files' load full reports and are slower. |
| limit | integer | Maximum rows to return (default 50, max 200). |
| region | string | Only this memory region, e.g. FLASH or RAM. |
| target* | string | Target name (see list_targets). |
| project* | string | The project: a public project's GitHub 'owner/repo' (no API key needed), or, with an API key, a project id or name from list_projects. |
| section | string | Only this section, e.g. .text (sections and symbols levels). |
| base_target | string | Compare with this other target of the project ('base' then names its build). |
| name_contains | string | Only sections, files or symbols whose name contains this text (case-insensitive). |
{
"type": "object",
"required": [
"project",
"target"
],
"properties": {
"base": {
"type": "string",
"description": "The older build, in the same forms. Default: the build 'head' was compared with when it was uploaded (for a pull request, the tip of its base branch); with 'base_target', the head of that target's default branch."
},
"head": {
"type": "string",
"description": "The newer build: a full commit SHA, a SHA prefix (7+ characters), or a branch or tag name. Default: the head of the default branch. For a pull request, its head commit SHA."
},
"level": {
"enum": [
"regions",
"sections",
"files",
"symbols"
],
"type": "string",
"description": "Detail to return (default 'regions'). 'symbols' and 'files' load full reports and are slower."
},
"limit": {
"type": "integer",
"maximum": 200,
"minimum": 1,
"description": "Maximum rows to return (default 50, max 200)."
},
"region": {
"type": "string",
"description": "Only this memory region, e.g. FLASH or RAM."
},
"target": {
"type": "string",
"description": "Target name (see list_targets)."
},
"project": {
"type": "string",
"description": "The project: a public project's GitHub 'owner/repo' (no API key needed), or, with an API key, a project id or name from list_projects."
},
"section": {
"type": "string",
"description": "Only this section, e.g. .text (sections and symbols levels)."
},
"base_target": {
"type": "string",
"description": "Compare with this other target of the project ('base' then names its build)."
},
"name_contains": {
"type": "string",
"description": "Only sections, files or symbols whose name contains this text (case-insensitive)."
}
},
"additionalProperties": false
}Get one build's whole symbol table as a file, to analyse with code (Python, pandas, DuckDB, jq) when the other tools cannot answer the question. Returns a one-time download, not the symbols: run the returned 'command' in a shell to save the file, then work on the file, so the table never enters the conversation. Needs a shell that can reach the MemBrowse server.
| Argument | Type | Description |
|---|---|---|
| ref | string | The build: a full commit SHA, a SHA prefix (7+ characters), or a branch or tag name. Default: the head of the default branch. |
| target* | string | Target name (see list_targets). |
| project* | string | The project: a public project's GitHub 'owner/repo' (no API key needed), or, with an API key, a project id or name from list_projects. |
{
"type": "object",
"required": [
"project",
"target"
],
"properties": {
"ref": {
"type": "string",
"description": "The build: a full commit SHA, a SHA prefix (7+ characters), or a branch or tag name. Default: the head of the default branch."
},
"target": {
"type": "string",
"description": "Target name (see list_targets)."
},
"project": {
"type": "string",
"description": "The project: a public project's GitHub 'owner/repo' (no API key needed), or, with an API key, a project id or name from list_projects."
}
},
"additionalProperties": false
}For one commit, show every target's memory change against its base build: per region, the largest section and source-file changes, budget alerts, and, for targets with benchmark metrics, each metric's change. For a pull request, pass the PR's head commit SHA: its base is the tip of the PR's base branch, so the result is the whole PR's change. Use it to review a pull request or check whether a commit grew memory or moved a benchmark.
| Argument | Type | Description |
|---|---|---|
| ref | string | The commit: a full commit SHA, a SHA prefix (7+ characters), or a branch or tag name. Default: the head of the default branch. For a pull request, its head commit SHA. |
| limit | integer | Section, file and metric changes per target (default 10, max 200). |
| target | string | Only this target (default: all). |
| project* | string | The project: a public project's GitHub 'owner/repo' (no API key needed), or, with an API key, a project id or name from list_projects. |
{
"type": "object",
"required": [
"project"
],
"properties": {
"ref": {
"type": "string",
"description": "The commit: a full commit SHA, a SHA prefix (7+ characters), or a branch or tag name. Default: the head of the default branch. For a pull request, its head commit SHA."
},
"limit": {
"type": "integer",
"maximum": 200,
"minimum": 1,
"description": "Section, file and metric changes per target (default 10, max 200)."
},
"target": {
"type": "string",
"description": "Only this target (default: all)."
},
"project": {
"type": "string",
"description": "The project: a public project's GitHub 'owner/repo' (no API key needed), or, with an API key, a project id or name from list_projects."
}
},
"additionalProperties": false
}List a target's commits, newest first, with each memory region's used size and the change that commit made. Use it to see how memory evolved and to find the commits that grew it (the largest positive 'delta'); page back with 'offset'.
| Argument | Type | Description |
|---|---|---|
| ref | string | Where to start: a full commit SHA, a SHA prefix (7+ characters), or a branch or tag name. Default: the head of the default branch. |
| limit | integer | Commits per page (default 30, max 200). |
| offset | integer | Commits to skip, for paging (default 0). |
| region | string | Only this memory region, e.g. FLASH or RAM. |
| target* | string | Target name (see list_targets). |
| project* | string | The project: a public project's GitHub 'owner/repo' (no API key needed), or, with an API key, a project id or name from list_projects. |
| only_changed | boolean | Skip commits whose build was identical to the previous one (default true). |
List a target's benchmarked commits, newest first, with each benchmark metric's value (cycle counts, timings, scores that CI attached) and its change from the previous commit that had the metric. Use it to see how a benchmark evolved and to find the commit that moved it; page back with 'offset'.
| Argument | Type | Description |
|---|---|---|
| ref | string | Where to start: a full commit SHA, a SHA prefix (7+ characters), or a branch or tag name. Default: the head of the default branch. |
| limit | integer | Commits per page (default 30, max 200). |
| offset | integer | Commits to skip, for paging (default 0). |
| target* | string | Target name (see list_targets). |
| metrics | array | Only these metrics (default: all of the target's metrics). |
| project* | string | The project: a public project's GitHub 'owner/repo' (no API key needed), or, with an API key, a project id or name from list_projects. |
{
"type": "object",
"required": [
"project",
"target"
],
"properties": {
"ref": {
"type": "string",
"description": "Where to start: a full commit SHA, a SHA prefix (7+ characters), or a branch or tag name. Default: the head of the default branch."
},
"limit": {
"type": "integer",
"maximum": 200,
"minimum": 1,
"description": "Commits per page (default 30, max 200)."
},
"offset": {
"type": "integer",
"minimum": 0,
"description": "Commits to skip, for paging (default 0)."
},
"target": {
"type": "string",
"description": "Target name (see list_targets)."
},
"metrics": {
"type": "array",
"items": {
"type": "string",
"minLength": 1
},
"maxItems": 50,
"minItems": 1,
"description": "Only these metrics (default: all of the target's metrics)."
},
"project": {
"type": "string",
"description": "The project: a public project's GitHub 'owner/repo' (no API key needed), or, with an API key, a project id or name from list_projects."
}
},
"additionalProperties": false
}List the MemBrowse projects this API key's account owns or is a member of. Needs an account API key. For a public project, skip this tool and call list_targets with its GitHub 'owner/repo'.
No arguments.
{
"type": "object",
"required": [],
"properties": {},
"additionalProperties": false
}List the largest symbols in one build of a target, optionally filtered by region, section, source file, object or name, or totalled per section, library, object, file or directory with 'group_by'. Use it to see what takes the space.
| Argument | Type | Description |
|---|---|---|
| ref | string | The build: a full commit SHA, a SHA prefix (7+ characters), or a branch or tag name. Default: the head of the default branch. |
| limit | integer | Maximum rows to return (default 50, max 200). |
| offset | integer | Rows to skip, for paging (default 0). |
| region | string | Only this memory region, e.g. FLASH or RAM. |
| target* | string | Target name (see list_targets). |
| details | boolean | Add address, binding, line, object and archive to each symbol (default false). |
| project* | string | The project: a public project's GitHub 'owner/repo' (no API key needed), or, with an API key, a project id or name from list_projects. |
| section | string |
List a project's build targets (firmware binaries) with the current used and limit size of each memory region at the head of the default branch.
| Argument | Type | Description |
|---|---|---|
| project* | string | The project: a public project's GitHub 'owner/repo' (no API key needed), or, with an API key, a project id or name from list_projects. |
{
"type": "object",
"required": [
"project"
],
"properties": {
"project": {
"type": "string",
"description": "The project: a public project's GitHub 'owner/repo' (no API key needed), or, with an API key, a project id or name from list_projects."
}
},
"additionalProperties": false
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 20, 2026.
{
"type": "object",
"required": [
"project",
"target"
],
"properties": {
"ref": {
"type": "string",
"description": "Where to start: a full commit SHA, a SHA prefix (7+ characters), or a branch or tag name. Default: the head of the default branch."
},
"limit": {
"type": "integer",
"maximum": 200,
"minimum": 1,
"description": "Commits per page (default 30, max 200)."
},
"offset": {
"type": "integer",
"minimum": 0,
"description": "Commits to skip, for paging (default 0)."
},
"region": {
"type": "string",
"description": "Only this memory region, e.g. FLASH or RAM."
},
"target": {
"type": "string",
"description": "Target name (see list_targets)."
},
"project": {
"type": "string",
"description": "The project: a public project's GitHub 'owner/repo' (no API key needed), or, with an API key, a project id or name from list_projects."
},
"only_changed": {
"type": "boolean",
"description": "Skip commits whose build was identical to the previous one (default true)."
}
},
"additionalProperties": false
}| Only this section, e.g. .bss. |
| group_by | string | Return totals per group instead of symbols. 'section' sizes whole sections, including string literals no symbol names ('unnamed_bytes'); name_contains then matches the section name. |
| file_contains | string | Only symbols whose source file path contains this text (case-insensitive). |
| name_contains | string | Only symbols whose name contains this text (case-insensitive). |
| object_contains | string | Only symbols whose archive or object file path contains this text (case-insensitive). |
{
"type": "object",
"required": [
"project",
"target"
],
"properties": {
"ref": {
"type": "string",
"description": "The build: a full commit SHA, a SHA prefix (7+ characters), or a branch or tag name. Default: the head of the default branch."
},
"limit": {
"type": "integer",
"maximum": 200,
"minimum": 1,
"description": "Maximum rows to return (default 50, max 200)."
},
"offset": {
"type": "integer",
"minimum": 0,
"description": "Rows to skip, for paging (default 0)."
},
"region": {
"type": "string",
"description": "Only this memory region, e.g. FLASH or RAM."
},
"target": {
"type": "string",
"description": "Target name (see list_targets)."
},
"details": {
"type": "boolean",
"description": "Add address, binding, line, object and archive to each symbol (default false)."
},
"project": {
"type": "string",
"description": "The project: a public project's GitHub 'owner/repo' (no API key needed), or, with an API key, a project id or name from list_projects."
},
"section": {
"type": "string",
"description": "Only this section, e.g. .bss."
},
"group_by": {
"enum": [
"section",
"library",
"object",
"file",
"directory"
],
"type": "string",
"description": "Return totals per group instead of symbols. 'section' sizes whole sections, including string literals no symbol names ('unnamed_bytes'); name_contains then matches the section name."
},
"file_contains": {
"type": "string",
"description": "Only symbols whose source file path contains this text (case-insensitive)."
},
"name_contains": {
"type": "string",
"description": "Only symbols whose name contains this text (case-insensitive)."
},
"object_contains": {
"type": "string",
"description": "Only symbols whose archive or object file path contains this text (case-insensitive)."
}
},
"additionalProperties": false
}