DAG Topological Scheduler

DAG Topological Scheduler MCP Connector for Claude

A+

Deterministic task scheduling and critical path analysis for multi-agent DAGs.

3 tools Official Updated Oct 1, 2026 Official Vinkius Partner

This MCP server provides a deterministic engine for multi-agent orchestration using Directed Acyclic Graphs (DAGs). It allows AI agents to calculate optimal task execution orders, identify the critical path, and manage parallel resource allocation. Using analyze_dag_structure, agents can validate graph integrity and detect cycles. The calculate_slack_and_bottlenecks tool identifies task flexibility and pinpoint bottlenecks, while simulate_agent_schedule calculates real-world makespan and parallelism efficiency for a given number of agents.

dagtopological-sortcritical-pathschedulingmulti-agent

3 tools expose this connector's capabilities to your AI agent.

analyze_dag_structure

Validates the integrity of the task graph and determines the core sequence and critical path

calculate_slack_and_bottlenecks

Identifies task flexibility and pinpoints the most sensitive tasks in the workflow

simulate_agent_schedule

Calculates the real-world execution timeline (makespan) and efficiency when limited agents are applied to the workload

See how to talk to your AI agent using DAG Topological Scheduler.

Analyze this task list for cycles and the critical path: [{'id': 'A', 'durationMs': 100, 'dependencies': []}, {'id': 'B', 'durationMs': 200, 'dependencies': ['A']}, {'id': 'C', 'durationMs': 150, 'dependencies': ['A']}]

The topological order is ['A', 'B', 'C'] or ['A', 'C', 'B']. The critical path is ['A', 'B'] with a duration of 300ms.

Simulate a schedule for these tasks with 2 agents: [{'id': 'T1', 'durationMs': 50, 'dependencies': []}, {'id': 'T2', 'durationMs': 50, 'dependencies': []}, {'id': 'T3', 'durationMs': 50, 'dependencies': ['T1', 'T2']}]

With 2 agents, the makespan is 100ms. T1 and T2 start at 0ms, and T3 starts at 50ms.

Find the bottleneck tasks for this DAG: [{'id': '1', 'durationMs': 10, 'dependencies': []}, {'id': '2', 'durationMs': 20, 'dependencies': ['1']}, {'id': '3', 'durationMs': 5, 'dependencies': ['1']}]

The bottleneck tasks are ['1', '2'] because they have zero slack time.

The server uses list scheduling to assign ready tasks to available agents, prioritizing tasks on the critical path to minimize total makespan.

Related Connectors