# @koreai/arch-mcp-tools

Source: https://glimind.com/server/npm%2F-koreai%2Farch-mcp-tools
Registry id: `npm/-koreai/arch-mcp-tools`
Status: unknown

## Install

```json
{
  "mcpServers": {
    "koreai-arch-mcp-tools": {
      "command": "npx",
      "args": [
        "-y",
        "@koreai/arch-mcp-tools"
      ]
    }
  }
}
```

Claude Code:

```sh
claude mcp add koreai-arch-mcp-tools -- npx -y @koreai/arch-mcp-tools
```

## Tools (56)

- `agent_tables`: Manage Agent Tables definitions and project-scoped rows
- `debug_diagnose_transcript`: Alias for transcript failure diagnosis
- `debug_explain_decision`: Explain agent decisions with context
- `debug_get_current_state`: Inspect agent context, gather progress, flow state
- `debug_get_errors`: Get errors, warnings, and escalations
- `debug_get_flow_graph`: View state machine graph (JSON or Mermaid)
- `debug_get_span_tree`: View hierarchical execution flow
- `debug_harness_logs`: Get CI execution logs
- `debug_lint_abl`: Run ABL repair and design lint checks
- `debug_list_active_sessions`: List observable sessions
- `debug_list_agents`: List available agents by domain
- `debug_load_agent`: Load an agent and create a debug session
- `debug_resume_session`: Reattach a known debug session on a fresh connection
- `debug_send_message`: Send a message to an agent
- `debug_session`: Subscribe/unsubscribe to session traces
- `debug_session_history`: Page durable Runtime sessions and persisted traces
- `debug_traces`: Search trace events (type, text, agent, error)
- `debug_why_transcript_failed`: Correlate transcript symptoms with ABL file/line causes
- `platform_agents`: List, get and create agents; guarded edit_dsl, prompt refs, legacy save_dsl
- `platform_alerts`: Discover metrics; create, update, test, inspect alert rules and evaluation history
- `platform_arch_auto_loop`: Drive project-scoped Arch Auto Loop repair workflows
- `platform_arch_sop`: Drive Studio Arch SOP-build sessions
- `platform_auth_profiles`: Manage profile metadata and start secure OAuth flows
- `platform_behavior_profiles`: Author and inspect project Behaviour Profiles and their agent usage
- `platform_config`: Manage project, project-effective Runtime, and LLM configuration
- `platform_connect`: Connect and authenticate to the platform
- `platform_deployments`: Manage typed deployments (list, get, create, promote, rollback, restore, retire)
- `platform_eval_evaluators`: Manage eval evaluators and templates
- `platform_eval_personas`: Manage and generate eval personas
- `platform_eval_runs`: Manage eval runs, preflight, cases, and heatmaps
- `platform_eval_scenarios`: Manage and generate eval scenarios
- `platform_eval_sets`: Manage eval sets and discover execution targets
- `platform_external_agents`: Register external agents, maintain endpoint/auth metadata, test connectivity
- `platform_feedback_templates`: List, get, create, update, inspect references and delete feedback templates
- `platform_guardrails`: Manage project policies and selected workspace providers
- `platform_import_export`: Import and export projects
- `platform_inbox`: Inspect and operate authorized project human tasks
- `platform_ingestion_pipelines`: Draft, validate, publish, and reindex SearchAI ingestion flows for KB
- `platform_insights_generator`: Preview and explicitly save synthetic Insights data on entitled projects
- `platform_integrations`: Create, update, test, and remove connector connections
- `platform_knowledge_bases`: Manage KBs, inline text and linked index sources; inspect health and rebuild
- `platform_mcp_servers`: Provision MCP servers; discover, import, and test tools
- `platform_memory`: Read project persistent-memory declarations from agent DSL
- `platform_omnichannel`: Manage channel policy and signed mobile webhook integrations, health, tests, rotation
- `platform_package_model`: Show compiler-visible agents, tools, constraints, refs, and diagnostics
- `platform_pipelines`: Author and operate durable project pipelines, runs, and custom output rows
- `platform_project_builder`: Discover provider contracts, inspect live dependencies/readiness, and plan
- `platform_project_builder_operations`: Continue durable operations and execute attempt-bound governed actions
- `platform_projects`: Manage projects (list, get, create, update, delete)
- `platform_prompt_library`: Author prompts and versions, test, promote, inspect references, manage lifecycle
- `platform_sdk_channels`: Create SDK keys and web/mobile/API channels
- `platform_tools`: Manage tools (list, get, create, update, delete, test)
- `platform_validate_package`: Validate a local package and optional import preview
- `platform_versions`: Manage immutable versions (list, get, publish, qualifications, audit, diff)
- `platform_workflows`: Create, update, publish, execute, and expose workflows as tools
- `platform_workspaces`: List, switch, and inspect active workspaces

Measured by Glimind (https://glimind.com): MCP handshake + tool list only, no tool is ever called. CC BY 4.0.
