diff --git a/docs/codex-claude-mcp-setup.md b/docs/codex-claude-mcp-setup.md
new file mode 100644
index 00000000..bb7b566a
--- /dev/null
+++ b/docs/codex-claude-mcp-setup.md
@@ -0,0 +1,117 @@
+# Codex and Claude MCP Setup
+
+## Goal
+
+Use MiroFish as a shared prediction engine through one MCP server so:
+
+- Codex can build and operate simulation flows
+- Claude can reason over the same simulation and report state
+- both tools stay aligned on one backend
+
+## Recommended Shape
+
+- Run the MiroFish backend locally
+- Run one MCP server against that backend
+- Point Codex and Claude at the same MCP server
+
+```text
+Codex ----\
+           >---- MiroFish MCP ---- MiroFish backend
+Claude ---/
+```
+
+## Start Order
+
+From [`/Users/al/Documents/CODEX/MiroFish`](/Users/al/Documents/CODEX/MiroFish):
+
+```bash
+npm run backend
+```
+
+In another terminal:
+
+```bash
+npm run mcp:stdio
+```
+
+If you prefer HTTP transport instead of stdio:
+
+```bash
+MIROFISH_BASE_URL=http://127.0.0.1:5001 npm run mcp:http
+```
+
+## Codex Connection
+
+Register the MCP server in Codex using the command below as the server command:
+
+```bash
+cd /Users/al/Documents/CODEX/MiroFish/backend && uv run python ../integrations/mcp/mirofish_mcp_server.py --transport stdio
+```
+
+Recommended environment:
+
+```text
+MIROFISH_BASE_URL=http://127.0.0.1:5001
+MIROFISH_HTTP_TIMEOUT=60
+```
+
+Use stdio for local Codex work unless you specifically need HTTP transport for shared remote access.
+
+## Claude Connection
+
+Register the same MCP server command in Claude's MCP settings:
+
+```bash
+cd /Users/al/Documents/CODEX/MiroFish/backend && uv run python ../integrations/mcp/mirofish_mcp_server.py --transport stdio
+```
+
+Use the same environment values:
+
+```text
+MIROFISH_BASE_URL=http://127.0.0.1:5001
+MIROFISH_HTTP_TIMEOUT=60
+```
+
+This keeps Codex and Claude on one shared tool surface and avoids separate wrapper logic.
+
+## MCP Tools You Can Rely On
+
+- `mirofish_health`
+- `mirofish_list_projects`
+- `mirofish_get_project`
+- `mirofish_build_graph`
+- `mirofish_get_graph_task`
+- `mirofish_create_simulation`
+- `mirofish_prepare_simulation`
+- `mirofish_prepare_status`
+- `mirofish_start_simulation`
+- `mirofish_stop_simulation`
+- `mirofish_simulation_status`
+- `mirofish_simulation_timeline`
+- `mirofish_interview_agents`
+- `mirofish_generate_report`
+- `mirofish_report_status`
+- `mirofish_get_report`
+- `mirofish_get_report_by_simulation`
+- `mirofish_chat_report`
+
+## Recommended Operator Flow
+
+1. Confirm backend health with `mirofish_health`.
+2. Choose an existing project with `mirofish_list_projects`.
+3. Build or rebuild the graph with `mirofish_build_graph`.
+4. Poll graph completion with `mirofish_get_graph_task`.
+5. Create a simulation with `mirofish_create_simulation`.
+6. Prepare the simulation with `mirofish_prepare_simulation`.
+7. Poll readiness with `mirofish_prepare_status`.
+8. Start the run with `mirofish_start_simulation`.
+9. Poll runtime with `mirofish_simulation_status`.
+10. Generate the report with `mirofish_generate_report`.
+11. Fetch the final output with `mirofish_get_report_by_simulation`.
+12. Ask follow-up questions with `mirofish_chat_report`.
+
+## Notes
+
+- Real simulation runs still require valid `LLM_API_KEY` and `ZEP_API_KEY`.
+- The MCP server is a control layer, not a replacement for the MiroFish backend.
+- n8n should call the backend directly for scheduled runs unless you explicitly want agent-mediated orchestration.
diff --git a/docs/integration-stack.md b/docs/integration-stack.md
new file mode 100644
index 00000000..6afa4ebf
--- /dev/null
+++ b/docs/integration-stack.md
@@ -0,0 +1,127 @@
+# MiroFish Integration Stack
+
+## Goal
+
+Use **MiroFish** as the prediction engine and simulation brain, with:
+
+- **Codex** as the builder and operator
+- **Claude** as the reasoning and synthesis layer
+- **n8n** as the automation layer
+- **MCP** as the shared tool access layer
+
+## Recommended Architecture
+
+```text
+Codex / Claude
+      |
+      v
+  MCP tools
+      |
+      v
+MiroFish backend API
+      |
+      +--> graph build
+      +--> simulation lifecycle
+      +--> interviews with agents
+      +--> report generation and chat
+      |
+      v
+  MiroFish frontend
+
+n8n
+  |
+  +--> calls MiroFish backend directly over HTTP
+  +--> or invokes the same lifecycle through the MCP server
+```
+
+## Roles
+
+### Codex
+
+- builds workflow glue, prompts, and wrappers
+- drives MiroFish through MCP tools
+- prepares projects, simulations, and reports
+
+### Claude
+
+- consumes simulation outputs and reports
+- performs deeper reasoning on what the simulation means
+- uses the same MCP tools so it sees the same state as Codex
+
+### n8n
+
+- schedules or triggers prediction runs
+- polls preparation/report task status
+- routes outputs to Slack, email, dashboards, or databases
+
+### MCP
+
+- standardizes access to MiroFish
+- gives Codex and Claude a stable tool layer
+- avoids coupling every agent directly to raw REST details
+
+## Added in This Setup
+
+- MCP server: [`integrations/mcp/mirofish_mcp_server.py`](/Users/al/Documents/CODEX/MiroFish/integrations/mcp/mirofish_mcp_server.py)
+- start scripts:
+  - [`scripts/start_mirofish_mcp_stdio.sh`](/Users/al/Documents/CODEX/MiroFish/scripts/start_mirofish_mcp_stdio.sh)
+  - [`scripts/start_mirofish_mcp_http.sh`](/Users/al/Documents/CODEX/MiroFish/scripts/start_mirofish_mcp_http.sh)
+- Codex and Claude setup guide:
+  - [`docs/codex-claude-mcp-setup.md`](/Users/al/Documents/CODEX/MiroFish/docs/codex-claude-mcp-setup.md)
+- n8n workflow template:
+  - [`integrations/n8n/mirofish_prediction_pipeline.json`](/Users/al/Documents/CODEX/MiroFish/integrations/n8n/mirofish_prediction_pipeline.json)
+- npm scripts:
+  - `npm run mcp:stdio`
+  - `npm run mcp:http`
+
+## MCP Tools Exposed
+
+- health
+- project list
+- project detail
+- graph build
+- graph task status
+- simulation create
+- simulation prepare
+- simulation prepare status
+- simulation start
+- simulation stop
+- simulation run status
+- simulation timeline
+- batch interviews
+- report generate
+- report generate status
+- report fetch
+- report fetch by simulation
+- report chat
+
+## n8n Automation Pattern
+
+Use HTTP Request nodes against the backend:
+
+1. `POST /api/graph/build`
+2. poll `GET /api/graph/task/{task_id}`
+3. `POST /api/simulation/create`
+4. `POST /api/simulation/prepare`
+5. poll `POST /api/simulation/prepare/status`
+6. `POST /api/simulation/start`
+7. poll `GET /api/simulation/{simulation_id}/run-status`
+8. `POST /api/report/generate`
+9. poll `POST /api/report/generate/status`
+10. `GET /api/report/by-simulation/{simulation_id}`
+
+An importable starter workflow is included at
+[`integrations/n8n/mirofish_prediction_pipeline.json`](/Users/al/Documents/CODEX/MiroFish/integrations/n8n/mirofish_prediction_pipeline.json).
+
+## Suggested Operating Mode
+
+- Let Codex build and evolve the simulation workflow.
+- Let Claude reason over the generated report and interview outputs.
+- Let n8n handle timed or event-driven execution.
+- Treat MiroFish as the core prediction substrate, not the orchestration system.
+
+## Notes
+
+- Real simulation quality depends on valid `LLM_API_KEY` and `ZEP_API_KEY`.
+- Local boot verification can use placeholder values, but real runs cannot.
+- Python 3.12 is required for this repo on this Mac.
diff --git a/integrations/mcp/mirofish_mcp_server.py b/integrations/mcp/mirofish_mcp_server.py
new file mode 100644
index 00000000..738b9058
--- /dev/null
+++ b/integrations/mcp/mirofish_mcp_server.py
@@ -0,0 +1,213 @@
+from __future__ import annotations
+
+import argparse
+import os
+from typing import Any
+
+import httpx
+from mcp.server.fastmcp import FastMCP
+
+
+DEFAULT_BASE_URL = os.environ.get("MIROFISH_BASE_URL", "http://127.0.0.1:5001")
+DEFAULT_TIMEOUT = float(os.environ.get("MIROFISH_HTTP_TIMEOUT", "60"))
+
+mcp = FastMCP(
+    name="MiroFish MCP",
+    instructions=(
+        "Use these tools to drive MiroFish as a prediction engine and simulation brain. "
+        "Create a project graph first, then create a simulation, prepare it, start it, "
+        "and generate or fetch reports."
+    ),
+)
+
+
+def _base_url() -> str:
+    return os.environ.get("MIROFISH_BASE_URL", DEFAULT_BASE_URL).rstrip("/")
+
+
+def _request(method: str, path: str, *, params: dict[str, Any] | None = None, json_body: dict[str, Any] | None = None) -> dict[str, Any]:
+    url = f"{_base_url()}{path}"
+    with httpx.Client(timeout=DEFAULT_TIMEOUT) as client:
+        response = client.request(method, url, params=params, json=json_body)
+        response.raise_for_status()
+        content_type = response.headers.get("content-type", "")
+        if "application/json" in content_type:
+            return response.json()
+        return {"success": True, "status_code": response.status_code, "text": response.text}
+
+
+@mcp.tool()
+def mirofish_health() -> dict[str, Any]:
+    return _request("GET", "/health")
+
+
+@mcp.tool()
+def mirofish_list_projects() -> dict[str, Any]:
+    return _request("GET", "/api/graph/project/list")
+
+
+@mcp.tool()
+def mirofish_get_project(project_id: str) -> dict[str, Any]:
+    return _request("GET", f"/api/graph/project/{project_id}")
+
+
+@mcp.tool()
+def mirofish_build_graph(payload: dict[str, Any]) -> dict[str, Any]:
+    return _request("POST", "/api/graph/build", json_body=payload)
+
+
+@mcp.tool()
+def mirofish_get_graph_task(task_id: str) -> dict[str, Any]:
+    return _request("GET", f"/api/graph/task/{task_id}")
+
+
+@mcp.tool()
+def mirofish_create_simulation(
+    project_id: str,
+    graph_id: str | None = None,
+    enable_twitter: bool = True,
+    enable_reddit: bool = True,
+) -> dict[str, Any]:
+    payload = {
+        "project_id": project_id,
+        "enable_twitter": enable_twitter,
+        "enable_reddit": enable_reddit,
+    }
+    if graph_id:
+        payload["graph_id"] = graph_id
+    return _request("POST", "/api/simulation/create", json_body=payload)
+
+
+@mcp.tool()
+def mirofish_prepare_simulation(
+    simulation_id: str,
+    entity_types: list[str] | None = None,
+    use_llm_for_profiles: bool = True,
+    parallel_profile_count: int = 5,
+    force_regenerate: bool = False,
+) -> dict[str, Any]:
+    payload: dict[str, Any] = {
+        "simulation_id": simulation_id,
+        "use_llm_for_profiles": use_llm_for_profiles,
+        "parallel_profile_count": parallel_profile_count,
+        "force_regenerate": force_regenerate,
+    }
+    if entity_types:
+        payload["entity_types"] = entity_types
+    return _request("POST", "/api/simulation/prepare", json_body=payload)
+
+
+@mcp.tool()
+def mirofish_prepare_status(task_id: str) -> dict[str, Any]:
+    return _request("POST", "/api/simulation/prepare/status", json_body={"task_id": task_id})
+
+
+@mcp.tool()
+def mirofish_start_simulation(
+    simulation_id: str,
+    platform: str = "parallel",
+    max_rounds: int | None = None,
+    enable_graph_memory_update: bool = False,
+    force: bool = False,
+) -> dict[str, Any]:
+    payload: dict[str, Any] = {
+        "simulation_id": simulation_id,
+        "platform": platform,
+        "enable_graph_memory_update": enable_graph_memory_update,
+        "force": force,
+    }
+    if max_rounds is not None:
+        payload["max_rounds"] = max_rounds
+    return _request("POST", "/api/simulation/start", json_body=payload)
+
+
+@mcp.tool()
+def mirofish_stop_simulation(simulation_id: str) -> dict[str, Any]:
+    return _request("POST", "/api/simulation/stop", json_body={"simulation_id": simulation_id})
+
+
+@mcp.tool()
+def mirofish_simulation_status(simulation_id: str, detailed: bool = False) -> dict[str, Any]:
+    suffix = "/run-status/detail" if detailed else "/run-status"
+    return _request("GET", f"/api/simulation/{simulation_id}{suffix}")
+
+
+@mcp.tool()
+def mirofish_simulation_timeline(simulation_id: str) -> dict[str, Any]:
+    return _request("GET", f"/api/simulation/{simulation_id}/timeline")
+
+
+@mcp.tool()
+def mirofish_interview_agents(
+    simulation_id: str,
+    interviews: list[dict[str, Any]],
+    platform: str = "parallel",
+) -> dict[str, Any]:
+    return _request(
+        "POST",
+        "/api/simulation/interview/batch",
+        json_body={
+            "simulation_id": simulation_id,
+            "platform": platform,
+            "interviews": interviews,
+        },
+    )
+
+
+@mcp.tool()
+def mirofish_generate_report(simulation_id: str, force_regenerate: bool = False) -> dict[str, Any]:
+    return _request(
+        "POST",
+        "/api/report/generate",
+        json_body={"simulation_id": simulation_id, "force_regenerate": force_regenerate},
+    )
+
+
+@mcp.tool()
+def mirofish_report_status(task_id: str) -> dict[str, Any]:
+    return _request("POST", "/api/report/generate/status", json_body={"task_id": task_id})
+
+
+@mcp.tool()
+def mirofish_get_report(report_id: str) -> dict[str, Any]:
+    return _request("GET", f"/api/report/{report_id}")
+
+
+@mcp.tool()
+def mirofish_get_report_by_simulation(simulation_id: str) -> dict[str, Any]:
+    return _request("GET", f"/api/report/by-simulation/{simulation_id}")
+
+
+@mcp.tool()
+def mirofish_chat_report(
+    simulation_id: str,
+    message: str,
+    chat_history: list[dict[str, str]] | None = None,
+) -> dict[str, Any]:
+    payload: dict[str, Any] = {
+        "simulation_id": simulation_id,
+        "message": message,
+    }
+    if chat_history:
+        payload["chat_history"] = chat_history
+    return _request("POST", "/api/report/chat", json_body=payload)
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description="Expose MiroFish backend APIs over MCP.")
+    parser.add_argument("--transport", choices=["stdio", "http"], default="stdio")
+    parser.add_argument("--host", default=os.environ.get("MIROFISH_MCP_HOST", "127.0.0.1"))
+    parser.add_argument("--port", type=int, default=int(os.environ.get("MIROFISH_MCP_PORT", "8000")))
+    args = parser.parse_args()
+
+    if args.transport == "http":
+        mcp.settings.host = args.host
+        mcp.settings.port = args.port
+        mcp.run(transport="streamable-http")
+        return
+
+    mcp.run()
+
+
+if __name__ == "__main__":
+    main()
diff --git a/integrations/n8n/mirofish_prediction_pipeline.json b/integrations/n8n/mirofish_prediction_pipeline.json
new file mode 100644
index 00000000..b4ca5347
--- /dev/null
+++ b/integrations/n8n/mirofish_prediction_pipeline.json
@@ -0,0 +1,650 @@
+{
+  "name": "MiroFish Prediction Pipeline",
+  "nodes": [
+    {
+      "parameters": {},
+      "id": "manual-trigger",
+      "name": "Manual Trigger",
+      "type": "n8n-nodes-base.manualTrigger",
+      "typeVersion": 1,
+      "position": [
+        -1360,
+        160
+      ]
+    },
+    {
+      "parameters": {
+        "assignments": {
+          "assignments": [
+            {
+              "id": "mirofish-base-url",
+              "name": "mirofishBaseUrl",
+              "value": "http://127.0.0.1:5001",
+              "type": "string"
+            },
+            {
+              "id": "project-id",
+              "name": "projectId",
+              "value": "proj_replace_me",
+              "type": "string"
+            },
+            {
+              "id": "graph-name",
+              "name": "graphName",
+              "value": "Prediction Brain Graph",
+              "type": "string"
+            },
+            {
+              "id": "chunk-size",
+              "name": "chunkSize",
+              "value": 500,
+              "type": "number"
+            },
+            {
+              "id": "chunk-overlap",
+              "name": "chunkOverlap",
+              "value": 50,
+              "type": "number"
+            },
+            {
+              "id": "enable-twitter",
+              "name": "enableTwitter",
+              "value": true,
+              "type": "boolean"
+            },
+            {
+              "id": "enable-reddit",
+              "name": "enableReddit",
+              "value": true,
+              "type": "boolean"
+            },
+            {
+              "id": "use-llm-profiles",
+              "name": "useLlmForProfiles",
+              "value": true,
+              "type": "boolean"
+            },
+            {
+              "id": "parallel-profile-count",
+              "name": "parallelProfileCount",
+              "value": 5,
+              "type": "number"
+            },
+            {
+              "id": "max-rounds",
+              "name": "maxRounds",
+              "value": 24,
+              "type": "number"
+            }
+          ]
+        },
+        "options": {}
+      },
+      "id": "set-inputs",
+      "name": "Set Inputs",
+      "type": "n8n-nodes-base.set",
+      "typeVersion": 3.4,
+      "position": [
+        -1140,
+        160
+      ]
+    },
+    {
+      "parameters": {
+        "method": "POST",
+        "url": "={{ $json.mirofishBaseUrl + '/api/graph/build' }}",
+        "sendBody": true,
+        "specifyBody": "json",
+        "jsonBody": "={{ { project_id: $json.projectId, graph_name: $json.graphName, chunk_size: $json.chunkSize, chunk_overlap: $json.chunkOverlap } }}",
+        "options": {}
+      },
+      "id": "build-graph",
+      "name": "Build Graph",
+      "type": "n8n-nodes-base.httpRequest",
+      "typeVersion": 4.2,
+      "position": [
+        -920,
+        160
+      ]
+    },
+    {
+      "parameters": {
+        "method": "GET",
+        "url": "={{ $('Set Inputs').item.json.mirofishBaseUrl + '/api/graph/task/' + $json.data.task_id }}",
+        "options": {}
+      },
+      "id": "poll-graph-task",
+      "name": "Poll Graph Task",
+      "type": "n8n-nodes-base.httpRequest",
+      "typeVersion": 4.2,
+      "position": [
+        -700,
+        160
+      ]
+    },
+    {
+      "parameters": {
+        "conditions": {
+          "string": [
+            {
+              "value1": "={{ $json.data.status }}",
+              "operation": "equal",
+              "value2": "completed"
+            }
+          ]
+        }
+      },
+      "id": "graph-complete",
+      "name": "Graph Complete?",
+      "type": "n8n-nodes-base.if",
+      "typeVersion": 2,
+      "position": [
+        -480,
+        160
+      ]
+    },
+    {
+      "parameters": {
+        "amount": 20,
+        "unit": "seconds"
+      },
+      "id": "wait-graph",
+      "name": "Wait Graph",
+      "type": "n8n-nodes-base.wait",
+      "typeVersion": 1.1,
+      "position": [
+        -480,
+        320
+      ]
+    },
+    {
+      "parameters": {
+        "method": "POST",
+        "url": "={{ $('Set Inputs').item.json.mirofishBaseUrl + '/api/simulation/create' }}",
+        "sendBody": true,
+        "specifyBody": "json",
+        "jsonBody": "={{ { project_id: $('Set Inputs').item.json.projectId, enable_twitter: $('Set Inputs').item.json.enableTwitter, enable_reddit: $('Set Inputs').item.json.enableReddit } }}",
+        "options": {}
+      },
+      "id": "create-simulation",
+      "name": "Create Simulation",
+      "type": "n8n-nodes-base.httpRequest",
+      "typeVersion": 4.2,
+      "position": [
+        -260,
+        80
+      ]
+    },
+    {
+      "parameters": {
+        "method": "POST",
+        "url": "={{ $('Set Inputs').item.json.mirofishBaseUrl + '/api/simulation/prepare' }}",
+        "sendBody": true,
+        "specifyBody": "json",
+        "jsonBody": "={{ { simulation_id: $json.data.simulation_id, use_llm_for_profiles: $('Set Inputs').item.json.useLlmForProfiles, parallel_profile_count: $('Set Inputs').item.json.parallelProfileCount } }}",
+        "options": {}
+      },
+      "id": "prepare-simulation",
+      "name": "Prepare Simulation",
+      "type": "n8n-nodes-base.httpRequest",
+      "typeVersion": 4.2,
+      "position": [
+        -40,
+        80
+      ]
+    },
+    {
+      "parameters": {
+        "method": "POST",
+        "url": "={{ $('Set Inputs').item.json.mirofishBaseUrl + '/api/simulation/prepare/status' }}",
+        "sendBody": true,
+        "specifyBody": "json",
+        "jsonBody": "={{ { task_id: $json.data.task_id } }}",
+        "options": {}
+      },
+      "id": "poll-prepare-status",
+      "name": "Poll Prepare Status",
+      "type": "n8n-nodes-base.httpRequest",
+      "typeVersion": 4.2,
+      "position": [
+        180,
+        80
+      ]
+    },
+    {
+      "parameters": {
+        "conditions": {
+          "string": [
+            {
+              "value1": "={{ $json.data.status }}",
+              "operation": "equal",
+              "value2": "ready"
+            }
+          ]
+        }
+      },
+      "id": "prepare-ready",
+      "name": "Prepare Ready?",
+      "type": "n8n-nodes-base.if",
+      "typeVersion": 2,
+      "position": [
+        400,
+        80
+      ]
+    },
+    {
+      "parameters": {
+        "amount": 30,
+        "unit": "seconds"
+      },
+      "id": "wait-prepare",
+      "name": "Wait Prepare",
+      "type": "n8n-nodes-base.wait",
+      "typeVersion": 1.1,
+      "position": [
+        400,
+        240
+      ]
+    },
+    {
+      "parameters": {
+        "method": "POST",
+        "url": "={{ $('Set Inputs').item.json.mirofishBaseUrl + '/api/simulation/start' }}",
+        "sendBody": true,
+        "specifyBody": "json",
+        "jsonBody": "={{ { simulation_id: $('Create Simulation').item.json.data.simulation_id, platform: 'parallel', max_rounds: $('Set Inputs').item.json.maxRounds, enable_graph_memory_update: false, force: false } }}",
+        "options": {}
+      },
+      "id": "start-simulation",
+      "name": "Start Simulation",
+      "type": "n8n-nodes-base.httpRequest",
+      "typeVersion": 4.2,
+      "position": [
+        620,
+        0
+      ]
+    },
+    {
+      "parameters": {
+        "method": "GET",
+        "url": "={{ $('Set Inputs').item.json.mirofishBaseUrl + '/api/simulation/' + $('Create Simulation').item.json.data.simulation_id + '/run-status' }}",
+        "options": {}
+      },
+      "id": "poll-run-status",
+      "name": "Poll Run Status",
+      "type": "n8n-nodes-base.httpRequest",
+      "typeVersion": 4.2,
+      "position": [
+        840,
+        0
+      ]
+    },
+    {
+      "parameters": {
+        "conditions": {
+          "string": [
+            {
+              "value1": "={{ $json.data.status }}",
+              "operation": "equal",
+              "value2": "completed"
+            }
+          ]
+        }
+      },
+      "id": "simulation-complete",
+      "name": "Simulation Complete?",
+      "type": "n8n-nodes-base.if",
+      "typeVersion": 2,
+      "position": [
+        1060,
+        0
+      ]
+    },
+    {
+      "parameters": {
+        "amount": 60,
+        "unit": "seconds"
+      },
+      "id": "wait-simulation",
+      "name": "Wait Simulation",
+      "type": "n8n-nodes-base.wait",
+      "typeVersion": 1.1,
+      "position": [
+        1060,
+        160
+      ]
+    },
+    {
+      "parameters": {
+        "method": "POST",
+        "url": "={{ $('Set Inputs').item.json.mirofishBaseUrl + '/api/report/generate' }}",
+        "sendBody": true,
+        "specifyBody": "json",
+        "jsonBody": "={{ { simulation_id: $('Create Simulation').item.json.data.simulation_id, force_regenerate: false } }}",
+        "options": {}
+      },
+      "id": "generate-report",
+      "name": "Generate Report",
+      "type": "n8n-nodes-base.httpRequest",
+      "typeVersion": 4.2,
+      "position": [
+        1280,
+        -80
+      ]
+    },
+    {
+      "parameters": {
+        "method": "POST",
+        "url": "={{ $('Set Inputs').item.json.mirofishBaseUrl + '/api/report/generate/status' }}",
+        "sendBody": true,
+        "specifyBody": "json",
+        "jsonBody": "={{ { task_id: $json.data.task_id } }}",
+        "options": {}
+      },
+      "id": "poll-report-status",
+      "name": "Poll Report Status",
+      "type": "n8n-nodes-base.httpRequest",
+      "typeVersion": 4.2,
+      "position": [
+        1500,
+        -80
+      ]
+    },
+    {
+      "parameters": {
+        "conditions": {
+          "string": [
+            {
+              "value1": "={{ $json.data.status }}",
+              "operation": "equal",
+              "value2": "completed"
+            }
+          ]
+        }
+      },
+      "id": "report-complete",
+      "name": "Report Complete?",
+      "type": "n8n-nodes-base.if",
+      "typeVersion": 2,
+      "position": [
+        1720,
+        -80
+      ]
+    },
+    {
+      "parameters": {
+        "amount": 30,
+        "unit": "seconds"
+      },
+      "id": "wait-report",
+      "name": "Wait Report",
+      "type": "n8n-nodes-base.wait",
+      "typeVersion": 1.1,
+      "position": [
+        1720,
+        80
+      ]
+    },
+    {
+      "parameters": {
+        "method": "GET",
+        "url": "={{ $('Set Inputs').item.json.mirofishBaseUrl + '/api/report/by-simulation/' + $('Create Simulation').item.json.data.simulation_id }}",
+        "options": {}
+      },
+      "id": "fetch-report",
+      "name": "Fetch Report",
+      "type": "n8n-nodes-base.httpRequest",
+      "typeVersion": 4.2,
+      "position": [
+        1940,
+        -160
+      ]
+    }
+  ],
+  "connections": {
+    "Manual Trigger": {
+      "main": [
+        [
+          {
+            "node": "Set Inputs",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Set Inputs": {
+      "main": [
+        [
+          {
+            "node": "Build Graph",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Build Graph": {
+      "main": [
+        [
+          {
+            "node": "Poll Graph Task",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Poll Graph Task": {
+      "main": [
+        [
+          {
+            "node": "Graph Complete?",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Graph Complete?": {
+      "main": [
+        [
+          {
+            "node": "Create Simulation",
+            "type": "main",
+            "index": 0
+          }
+        ],
+        [
+          {
+            "node": "Wait Graph",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Wait Graph": {
+      "main": [
+        [
+          {
+            "node": "Poll Graph Task",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Create Simulation": {
+      "main": [
+        [
+          {
+            "node": "Prepare Simulation",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Prepare Simulation": {
+      "main": [
+        [
+          {
+            "node": "Poll Prepare Status",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Poll Prepare Status": {
+      "main": [
+        [
+          {
+            "node": "Prepare Ready?",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Prepare Ready?": {
+      "main": [
+        [
+          {
+            "node": "Start Simulation",
+            "type": "main",
+            "index": 0
+          }
+        ],
+        [
+          {
+            "node": "Wait Prepare",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Wait Prepare": {
+      "main": [
+        [
+          {
+            "node": "Poll Prepare Status",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Start Simulation": {
+      "main": [
+        [
+          {
+            "node": "Poll Run Status",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Poll Run Status": {
+      "main": [
+        [
+          {
+            "node": "Simulation Complete?",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Simulation Complete?": {
+      "main": [
+        [
+          {
+            "node": "Generate Report",
+            "type": "main",
+            "index": 0
+          }
+        ],
+        [
+          {
+            "node": "Wait Simulation",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Wait Simulation": {
+      "main": [
+        [
+          {
+            "node": "Poll Run Status",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Generate Report": {
+      "main": [
+        [
+          {
+            "node": "Poll Report Status",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Poll Report Status": {
+      "main": [
+        [
+          {
+            "node": "Report Complete?",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Report Complete?": {
+      "main": [
+        [
+          {
+            "node": "Fetch Report",
+            "type": "main",
+            "index": 0
+          }
+        ],
+        [
+          {
+            "node": "Wait Report",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    },
+    "Wait Report": {
+      "main": [
+        [
+          {
+            "node": "Poll Report Status",
+            "type": "main",
+            "index": 0
+          }
+        ]
+      ]
+    }
+  },
+  "active": false,
+  "settings": {},
+  "pinData": {},
+  "meta": {
+    "templateCredsSetupCompleted": false
+  },
+  "versionId": "1"
+}
diff --git a/package.json b/package.json
index 3baf424a..1a165054 100644
--- a/package.json
+++ b/package.json
@@ -8,6 +8,8 @@
     "setup:all": "npm run setup && npm run setup:backend",
     "dev": "concurrently --kill-others -n \"backend,frontend\" -c \"green,cyan\" \"npm run backend\" \"npm run frontend\"",
     "backend": "cd backend && uv run python run.py",
+    "mcp:stdio": "cd backend && uv run python ../integrations/mcp/mirofish_mcp_server.py --transport stdio",
+    "mcp:http": "cd backend && uv run python ../integrations/mcp/mirofish_mcp_server.py --transport http",
     "frontend": "cd frontend && npm run dev",
     "build": "cd frontend && npm run build"
   },
diff --git a/scripts/start_mirofish_mcp_http.sh b/scripts/start_mirofish_mcp_http.sh
new file mode 100755
index 00000000..0ad595b3
--- /dev/null
+++ b/scripts/start_mirofish_mcp_http.sh
@@ -0,0 +1,4 @@
+#!/bin/sh
+set -eu
+cd /Users/al/Documents/CODEX/MiroFish/backend
+uv run python ../integrations/mcp/mirofish_mcp_server.py --transport http
diff --git a/scripts/start_mirofish_mcp_stdio.sh b/scripts/start_mirofish_mcp_stdio.sh
new file mode 100755
index 00000000..e39d47e1
--- /dev/null
+++ b/scripts/start_mirofish_mcp_stdio.sh
@@ -0,0 +1,4 @@
+#!/bin/sh
+set -eu
+cd /Users/al/Documents/CODEX/MiroFish/backend
+uv run python ../integrations/mcp/mirofish_mcp_server.py --transport stdio