// SCAN REPORT

Swanlab

11 tools. 0 can modify or destroy data without limits.

Read-only server. Low risk, but rate limits prevent runaway API costs.

Last updated: 12 Jun 2026

0 can modify or destroy data

11 read-only

11 tools total

Community server · catalogue entry verified 12/06/2026

How to control Swanlab ↓

TOOL MAP

What Swanlab exposes to your agents

Read (11) Write / Execute (0) Destructive / Financial (0)

Running more than one server? Check your whole stack →

THE RISK

What can go wrong

Even read-only tools carry cost. An agent in a retry loop can make thousands of API calls per minute, exhausting rate limits and running up bills.

POLICY

How to control Swanlab

PolicyLayer is an MCP gateway — it sits between your AI agents and Swanlab, and nothing reaches the server without passing your rules. These are the rules we recommend:

Cap read operations

{
  "swanlab_get_project": {
    "limits": [
      {
        "counter": "swanlab_get_project_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register Swanlab — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON SWANLAB →

Free to start. No card required.

FULL CATALOGUE

All 11 Swanlab tools

READ 11 tools

Read swanlab_get_project Get detailed information about a specific project. 获取指定项目的详细信息。 Read swanlab_get_run Get detailed information about a specific run (experiment). 获取指定实验的详细信息。 Read swanlab_get_run_config Get the configuration (config) for a specific run (experiment). 获取实验的配置信息。 Read swanlab_get_run_metadata Get the environment metadata for a specific run (experiment). Read swanlab_get_run_metrics Get metric data for a run (experiment). Returns a list of metric records. Read swanlab_get_run_requirements Get the Python requirements for a specific run (experiment). 获取实验的 Python 依赖信息。 Read swanlab_get_workspace Get detailed information about a specific workspace. 获取指定工作空间的详细信息。 Read swanlab_list_projects List all projects with optional filtering by workspace, sort, and search. Read swanlab_list_run_metric_keys List all available metric keys for a run (experiment). Read swanlab_list_runs List all runs (experiments) in a project with optional filtering. Read swanlab_list_workspaces List all workspaces accessible to the current user. 工作空间是项目的集合，对应一个研发团队。

FAQ

Questions about Swanlab

Is the Swanlab MCP server safe to use without restrictions? +

The Swanlab server is primarily read-only with 11 read tools. While it cannot modify data, an agent in a retry loop can make thousands of API calls per minute, exhausting rate limits and running up costs. Rate limiting is still recommended.

How many tools does the Swanlab MCP server expose? +

11 tools across 1 categories: Read. 11 are read-only. 0 can modify, create, or delete data.

How do I enforce a policy on Swanlab? +

Register the Swanlab MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every Swanlab tool call.

Deterministic rules across all 11 Swanlab tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON SWANLAB →

Free to start. No card required.

11 Swanlab tools catalogued and risk-classified — across an index of 43,000+ MCP servers.