// SCAN REPORT

Langfuse Mcp Python

40 tools. 22 can modify or destroy data without limits.

7 destructive tools with no built-in limits. Policy required.

Last updated: 30 Jun 2026

22 can modify or destroy data

18 read-only

40 tools total

Community server · catalogue entry verified 30/06/2026

How to control Langfuse Mcp Python ↓

TOOL MAP

What Langfuse Mcp Python exposes to your agents

Read (18) Write / Execute (15) Destructive / Financial (7)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Langfuse Mcp Python tools

Destructive · High delete_annotation_queue The tool permanently deletes an annotation queue, which cannot be undone. Destructive · High delete_blob_storage_integration The tool performs an irreversible deletion of a blob storage integration configuration. Destructive · High delete_dataset This tool permanently removes data without the possibility of reversal. Destructive · High delete_model This tool permanently removes a model configuration with no undo mechanism.

22 of Langfuse Mcp Python's 40 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Langfuse Mcp Python

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

Deny destructive operations

{
  "delete_annotation_queue": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Blocked by default. Requires approval."
      }
    ]
  }
}

Destructive tools should never be available to autonomous agents without human approval.

Rate limit write operations

{
  "add_comment": {
    "limits": [
      {
        "counter": "add_comment_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "analyze_performance": {
    "limits": [
      {
        "counter": "analyze_performance_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 Langfuse Mcp Python — 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 LANGFUSE MCP PYTHON →

Instant setup, no code required.

FULL CATALOGUE

All 40 Langfuse Mcp Python tools

DESTRUCTIVE 7 tools

Destructive delete_annotation_queue Delete an annotation queue. Destructive delete_blob_storage_integration Delete a blob storage integration by ID. Destructive delete_dataset Delete a dataset and all its items. Destructive delete_model Delete a custom model configuration. Destructive delete_project Delete a project by ID (requires organization-scoped API key). Destructive delete_prompt Delete a specific prompt version or all versions. Destructive delete_trace Delete a trace and all its associated observations.

WRITE 15 tools

Write add_comment Add a comment to a trace or observation. Write create_annotation_queue Create a new annotation queue. Write create_dataset Create a new dataset for experiments and evaluations. Write create_dataset_item Add an item to a dataset. Write create_model Create a custom model configuration with pricing. Write create_project Create a new project (requires organization-scoped API key). Write create_prompt Create a new prompt or new version of existing prompt. Write resolve_queue_item Mark an annotation queue item as resolved (status=COMPLETED). Write submit_score Submit an evaluation score for a trace or observation. Write update_annotation_queue Update annotation queue properties. Write update_dataset Update dataset properties. Write update_project Update a project by ID (requires organization-scoped API key). Write update_trace Update trace metadata, tags, or other properties. Write upsert_blob_storage_integration Create or update a blob storage integration (requires organization-scoped API key). Write upsert_llm_connection Create or update an LLM connection (upsert by provider).

READ 18 tools

Read analyze_performance Analyze agent performance by aggregating observations directly (no Metrics API). Read get_annotation_queues List all annotation queues for human review. Read get_blob_storage_integration_status Get sync status for a blob storage integration by ID. Read get_blob_storage_integrations List all blob storage integrations (requires organization-scoped API key). Read get_comments Retrieve comments for a trace or observation. Read get_cost_metrics Query Langfuse cost, token, latency, and usage analytics via the Metrics API. Read get_datasets Retrieve datasets and dataset items for experiments and evaluations. Read get_llm_connections List all LLM connections for the project. Read get_metrics Get aggregated metrics and analytics from Langfuse. Read get_models Retrieve model configurations and pricing information. Read get_projects Get project(s) associated with the API key. Read get_prompts Retrieve and manage prompts from Langfuse. Read get_queue_items Get items in an annotation queue. Read get_score_configs Retrieve score configuration definitions. Includes available scores, their types, and descriptions. and score Read get_scores Retrieve evaluation scores for traces. Read get_sessions Retrieve multi-turn agent sessions. Read get_trace Retrieve detailed execution trace with all observations and metrics. Read watch_agents Monitor active agents in real-time with performance metrics.

EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

FAQ

Questions about Langfuse Mcp Python

Can an AI agent delete data through the Langfuse Mcp Python MCP server? +

Yes. The Langfuse Mcp Python server exposes 7 destructive tools including delete_annotation_queue, delete_blob_storage_integration, delete_dataset. These permanently remove resources with no undo. PolicyLayer blocks destructive tools by default so they never reach the upstream server.

How do I prevent bulk modifications through Langfuse Mcp Python? +

The Langfuse Mcp Python server has 15 write tools including add_comment, create_annotation_queue, create_dataset. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Langfuse Mcp Python.

How many tools does the Langfuse Mcp Python MCP server expose? +

40 tools across 3 categories: Destructive, Read, Write. 18 are read-only. 22 can modify, create, or delete data.

How do I enforce a policy on Langfuse Mcp Python? +

Register the Langfuse Mcp Python 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 Langfuse Mcp Python tool call.

Deterministic rules across all 40 Langfuse Mcp Python tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON LANGFUSE MCP PYTHON →

Instant setup, no code required.

40 Langfuse Mcp Python tools catalogued and risk-classified — across an index of 43,000+ MCP servers.