Skip to main content
Glama
agloop-mcp

agloop-mcp

by zebbern

agloop-mcp

MCP server providing 14 agent introspection tools for AgLoop — the VS Code Copilot custom agent framework.

Quick Start

Add to your .vscode/mcp.json:

For claude desktop or cli users use mcpservers instead of servers

{
  "servers": {
    "agloop": {
      "command": "uvx",
      "args": ["agloop-mcp"]
    }
  }
}

Or run directly:

uvx agloop-mcp

Related MCP server: mcp-tool-server

Tools

Read Tools (10)

Tool

Description

agloop_get_state

Full AgLoop state: phase, tasks, iteration, compaction context

agloop_get_task

Single task details by ID

agloop_list_tasks

All tasks with optional status/dependency filter

agloop_get_logs

Execution log entries with agent/task/limit filters

agloop_get_plan

Current plan (YAML or Markdown)

agloop_get_compaction_context

Compaction recovery context

agloop_search_logs

Full-text search through execution logs

agloop_get_agent_info

Agent definition file (.agent.md) content

agloop_get_next_task

Topological-sort next task recommendation

agloop_list_agents

All available agent names

Write Tools (4)

Tool

Description

agloop_update_task

Update task status and result log

agloop_append_log

Append structured entry to execution log

agloop_create_checkpoint

Snapshot current state for recovery

agloop_set_phase

Transition to a new phase

Configuration

The server uses the current working directory as the workspace root. Override with:

uvx agloop-mcp --workspace /path/to/project

State Files

The server reads/writes from .agloop/ in the workspace root:

.agloop/
├── state.json        # Phase, tasks, iteration
├── log.json          # Execution log (JSON Lines)
├── plan.yaml         # Current plan
└── checkpoints/      # State snapshots

Requirements

  • Python 3.11+

  • No additional dependencies beyond mcp

License

MIT

Available Tools

14 tools
agloop_append_logB

Append a structured entry to the execution log. COORDINATOR ONLY.

Inputs: agent (required), action (required), phase (required), status (required), +3 more. Returns: result

agloop_create_checkpointC

Create a checkpoint of the current state for recovery. COORDINATOR ONLY.

Inputs: label, trigger. Returns: result

agloop_get_agent_infoA

Read the definition file (.agent.md) for a specific agent. Returns the full agent prompt including frontmatter.

Inputs: agent_name (required). Returns: result

agloop_get_compaction_contextA

Get compaction recovery context: compaction count, pending decision, last delegation, recovery hint. Critical after VS Code context compaction.

Inputs: no named parameters. Returns: result

agloop_get_logsA

Read execution log entries. Supports filtering by agent, task, and limiting results.

Inputs: limit, agent, task_id. Returns: result

agloop_get_next_taskA

Compute the next task using topological sort. Returns the recommended next task and reasoning.

Inputs: no named parameters. Returns: result

agloop_get_planA

Read the current plan (plan.yaml or plan.md). Returns raw file contents.

Inputs: no named parameters. Returns: result

agloop_get_stateA

Read the full AgLoop state including phase, tasks, iteration, and compaction context. Returns null if no active loop.

Inputs: no named parameters. Returns: result

agloop_get_taskA

Get details for a single task by ID. Returns task status, dependencies, acceptance criteria, and result log.

Inputs: task_id (required). Returns: result

agloop_list_agentsA

List all available agent names in .github/agents/.

Inputs: no named parameters. Returns: result

agloop_list_tasksB

List all tasks with optional status filter. Returns array of task objects.

Inputs: status, depends_on. Returns: result

agloop_search_logsB

Full-text search through execution logs. Case-insensitive search across agent, action, task_id, phase, input/output summaries.

Inputs: query (required). Returns: result

agloop_set_phaseD

Transition to a new phase. COORDINATOR ONLY.

Inputs: phase (required). Returns: result

agloop_update_taskA

Update a task's status and/or result log. Auto-sets timestamps. COORDINATOR ONLY.

Inputs: task_id (required), status, result_log. Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.5/5.0

    Scored across 14 tools

    Disambiguation5/5

    Each tool targets a distinct aspect of the AgLoop workflow: logging, state management, agent info, task operations, and phase transitions. There is no overlap or ambiguity between tools.

    Naming Consistency5/5

    All tools follow a consistent 'agloop_verb_noun' pattern with appropriate verbs (append, create, get, list, search, set, update) and clear nouns, making the naming predictable.

    Tool Count5/5

    With 14 tools, the server covers the needed operations for a workflow management system without being excessive. The count is well-scoped for its purpose.

    Completeness4/5

    The tool set covers core workflow operations such as reading state, managing tasks and agents, logging, and phases. A minor gap is the lack of task creation or deletion tools, but tasks appear to be defined externally, so the surface is largely complete.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      C
      maintenance
      An MCP server that provides AI coding agents with AST-accurate, context-budget-aware codebase querying, safety gates, and team policy integration via structured tools and a local plugin layer.
      121 npm
      4
      MIT
    • F
      license
      Not graded
      quality
      D
      maintenance
      A custom MCP server with 6 utility tools (file search, file reading, math calculation, JSON formatting, time query, system info) that demonstrates MCP protocol workflow and integrates with LangChain agents.
      -