### Install context-portal-mcp Package

Source: https://github.com/greatscottymac/context-portal/blob/main/v0.2.4_UPDATE_GUIDE.md

Use this command to install the new version of the context-portal-mcp package. `uv pip install` is recommended.

```bash
uv pip install context-portal-mcp
```

```bash
# Alternatively, using pip:
# pip install context-portal-mcp
```

--------------------------------

### Install Dependencies with uv

Source: https://github.com/greatscottymac/context-portal/blob/main/CONTRIBUTING.md

Install project dependencies from requirements.txt using uv after activating the virtual environment.

```bash
uv pip install -r requirements.txt
```

--------------------------------

### Install Dependencies with pip

Source: https://github.com/greatscottymac/context-portal/blob/main/CONTRIBUTING.md

Install project dependencies from requirements.txt using pip after activating the virtual environment.

```bash
pip install -r requirements.txt
```

--------------------------------

### Install Context Portal MCP v0.3.7 via uvx

Source: https://github.com/greatscottymac/context-portal/blob/main/RELEASE_NOTES.md

Use this command to install version 0.3.7 of the Context Portal MCP using uvx.

```bash
uvx --from git+https://github.com/GreatScottyMac/context-portal.git conport-mcp --mode stdio
```

--------------------------------

### Installation and Server Configuration

Source: https://context7.com/greatscottymac/context-portal/llms.txt

Instructions on how to install and configure the Context Portal MCP (ConPort) server, including MCP client configuration and running the server in STDIO or HTTP mode.

```APIDOC
## Installation and Server Configuration

ConPort can be installed via uvx (recommended) or pip, and runs as an MCP server that communicates with AI assistants through STDIO or HTTP protocols.

### MCP Client Configuration (using uvx)

```json
{
  "mcpServers": {
    "conport": {
      "command": "uvx",
      "args": [
        "--from",
        "context-portal-mcp",
        "conport-mcp",
        "--mode",
        "stdio",
        "--workspace_id",
        "/path/to/your/project",
        "--log-file",
        "./logs/conport.log",
        "--log-level",
        "INFO"
      ]
    }
  }
}
```

### Installation via pip

```bash
pip install context-portal-mcp
```

### Running the Server

#### STDIO Mode

```bash
conport-mcp --mode stdio --workspace_id "/path/to/project"
```

#### HTTP Mode

```bash
conport-mcp --mode http --host 127.0.0.1 --port 8000
```

*Note: HTTP mode is accessible at `http://localhost:8000/mcp`.*
```

--------------------------------

### Install Context Portal MCP v0.3.10 via uvx

Source: https://github.com/greatscottymac/context-portal/blob/main/RELEASE_NOTES.md

Use this command to install version 0.3.10 of the Context Portal MCP using uvx.

```bash
uvx --from git+https://github.com/GreatScottyMac/context-portal.git@v0.3.10 conport-mcp --mode stdio
```

--------------------------------

### Draft Minimal Public Quickstart Document

Source: https://github.com/greatscottymac/context-portal/blob/main/conport-custom-instructions/mem4sprint.schema_and_templates.md

Creates a draft for a minimal public quickstart document. Replace `<ABS_PATH>` with the workspace ID. The document is tagged as 'public_doc' and 'quickstart', with visibility set to 'public' and status 'DRAFT'.

```javascript
mcp0_log_custom_data({ workspace_id: "<ABS_PATH>", category: "artifacts", key: "quickstart", value: { artifact_kind: "doc", title: "Quickstart", tags: ["public_doc","quickstart"], visibility: "public", publish_status: "DRAFT" } })
```

--------------------------------

### Install Context Portal MCP v0.3.8 via uvx

Source: https://github.com/greatscottymac/context-portal/blob/main/RELEASE_NOTES.md

Use this command to install version 0.3.8 of the Context Portal MCP using uvx.

```bash
uvx --from git+https://github.com/GreatScottyMac/context-portal.git@v0.3.8 conport-mcp --mode stdio
```

--------------------------------

### Install Context Portal MCP v0.3.9 via uvx

Source: https://github.com/greatscottymac/context-portal/blob/main/RELEASE_NOTES.md

Use this command to install version 0.3.9 of the Context Portal MCP using uvx.

```bash
uvx --from git+https://github.com/GreatScottyMac/context-portal.git@v0.3.9 conport-mcp --mode stdio
```

--------------------------------

### Install Context Portal MCP v0.3.11 via uvx

Source: https://github.com/greatscottymac/context-portal/blob/main/RELEASE_NOTES.md

Use this command to install version 0.3.11 of the Context Portal MCP using uvx.

```bash
uvx --from git+https://github.com/GreatScottyMac/context-portal.git@v0.3.11 conport-mcp --mode stdio
```

--------------------------------

### Install Context Portal MCP v0.3.12 via uvx

Source: https://github.com/greatscottymac/context-portal/blob/main/RELEASE_NOTES.md

Use this command to install version 0.3.12 of the Context Portal MCP using uvx.

```bash
uvx --from git+https://github.com/GreatScottyMac/context-portal.git@v0.3.12 conport-mcp --mode stdio
```

--------------------------------

### Draft Quickstart Document

Source: https://github.com/greatscottymac/context-portal/blob/main/conport-custom-instructions/mem4sprint.schema_and_templates.md

Logs a custom data entry for a public quickstart document in draft status.

```APIDOC
## POST /api/mcp/log_custom_data

### Description
Logs a custom data entry for a public quickstart document in draft status.

### Method
POST

### Endpoint
/api/mcp/log_custom_data

### Parameters
#### Request Body
- **workspace_id** (string) - Required - The absolute workspace path.
- **category** (string) - Required - Set to "artifacts".
- **key** (string) - Required - Set to "quickstart".
- **value** (object) - Required - The quickstart document content.
  - **artifact_kind** (string) - Required - Set to "doc".
  - **title** (string) - Required - The title of the quickstart (e.g., "Quickstart").
  - **tags** (array) - Required - Array of tags, including "public_doc" and "quickstart".
  - **visibility** (string) - Required - Set to "public".
  - **publish_status** (string) - Required - Set to "DRAFT".
```

--------------------------------

### Install Context Portal MCP v0.3.13 via uvx

Source: https://github.com/greatscottymac/context-portal/blob/main/RELEASE_NOTES.md

Use this command to install version 0.3.13 of the Context Portal MCP using uvx.

```bash
uvx --from git+https://github.com/GreatScottyMac/context-portal.git@v0.3.13 conport-mcp --mode stdio
```

--------------------------------

### Get CLI Help

Source: https://github.com/greatscottymac/context-portal/blob/main/README.md

Displays help information for the Context Portal MCP main script when run via uv.

```bash
uv run python src/context_portal_mcp/main.py --help
```

--------------------------------

### Install and Run ConPort Server

Source: https://context7.com/greatscottymac/context-portal/llms.txt

Install ConPort using pip and run it in either STDIO mode for IDE integration or HTTP mode for broader accessibility.

```bash
# Alternative: Install via pip
pip install context-portal-mcp

# Run in STDIO mode
conport-mcp --mode stdio --workspace_id "/path/to/project"

# Run in HTTP mode (accessible at http://localhost:8000/mcp)
conport-mcp --mode http --host 127.0.0.1 --port 8000
```

--------------------------------

### Install Context Portal MCP v0.3.13 via pip

Source: https://github.com/greatscottymac/context-portal/blob/main/RELEASE_NOTES.md

Use this command to install version 0.3.13 of the Context Portal MCP using pip.

```bash
pip install context-portal-mcp
```

--------------------------------

### Install New Package with uv pip

Source: https://github.com/greatscottymac/context-portal/blob/main/v0.2.4_UPDATE_GUIDE.md

Use uv pip to install the latest version of the context-portal-mcp package. This is a common step in both automatic and manual migration paths.

```bash
uv pip install context-portal-mcp
```

--------------------------------

### Starting LLM Agent Interaction

Source: https://github.com/greatscottymac/context-portal/blob/main/README.md

To ensure correct ConPort initialization and context loading, start interactions with a clear directive, especially in interfaces that may not strictly adhere to custom instructions on the first message.

```text
Initialize according to custom instructions.
```

--------------------------------

### Minimal Auto-Detection Configuration

Source: https://github.com/greatscottymac/context-portal/blob/main/UNIVERSAL_WORKSPACE_DETECTION.md

Use this configuration for minimal setup where auto-detection is enabled by default.

```json
{
  "mcpServers": {
    "conport": {
      "command": "uvx",
      "args": [
        "--from", "context-portal-mcp",
        "conport-mcp",
        "--mode", "stdio",
        "--log-level", "INFO"
      ]
    }
  }
}
```

--------------------------------

### Custom Start Path for Workspace Detection

Source: https://github.com/greatscottymac/context-portal/blob/main/UNIVERSAL_WORKSPACE_DETECTION.md

Use the `--workspace-search-start` flag to specify a custom directory for initiating the workspace search, useful in deep launch scenarios.

```bash
conport-mcp --mode stdio --workspace-search-start ../../
```

--------------------------------

### Turn Start Fetch Recipes

Source: https://github.com/greatscottymac/context-portal/blob/main/conport-custom-instructions/mem4sprint.md

Provides recipes for fetching requirements, architecture plans, and top recent decisions for a sprint using ConPort search functions.

```yaml
turn_start_fetch_recipes:
    - name: "fetch_requirements_architecture_plan"
      calls: |
        mcp0_search_custom_data_value_fts({ workspace_id: "<ABS_PATH>", query_term: "value_text:\"artifact_kind:doc\" value_text:\"tag:requirements\"", limit: 5 })
        mcp0_search_custom_data_value_fts({ workspace_id: "<ABS_PATH>", query_term: "value_text:\"artifact_kind:doc\" value_text:\"tag:architecture\"", limit: 5 })
        mcp0_search_custom_data_value_fts({ workspace_id: "<ABS_PATH>", query_term: "value_text:\"artifact_kind:doc\" value_text:\"tag:sprint_plan\" value_text:\"sprint_id:S-YYYY.MM\"", limit: 5 })
    - name: "fetch_top_recent_decisions_for_sprint"
      calls: |
        mcp0_search_decisions_fts({ workspace_id: "<ABS_PATH>", query_term: "tags:\"S-YYYY.MM\"", limit: 5 })
```

--------------------------------

### Get System Patterns

Source: https://context7.com/greatscottymac/context-portal/llms.txt

Retrieves system patterns with optional tag filtering.

```APIDOC
## GET /get_system_patterns

### Description
Retrieves system patterns, with optional filtering by tags.

### Method
GET

### Endpoint
/get_system_patterns

### Query Parameters
- **workspace_id** (string) - Required - The ID of the workspace.
- **tags_filter_include_any** (array of strings) - Optional - Filters patterns to include any of the specified tags.
- **limit** (integer) - Optional - The maximum number of patterns to return.

### Response
#### Success Response (200)
- **id** (integer) - The unique identifier of the pattern.
- **name** (string) - The name of the pattern.
- **description** (string) - The description of the pattern.
- **tags** (array of strings) - Tags associated with the pattern.

#### Response Example
```json
[
    {
        "id": 1,
        "name": "Repository Pattern",
        "description": "Data access layer uses repository pattern...",
        "tags": ["architecture", "data-access"]
    },
    {
        "id": 2,
        "name": "CQRS for Orders",
        "description": "Order processing uses Command Query Responsibility Segregation...",
        "tags": ["architecture", "orders"]
    }
]
```
```

--------------------------------

### MCP Client Configuration with uvx

Source: https://context7.com/greatscottymac/context-portal/llms.txt

Configure the MCP client to connect to the ConPort server using uvx. Ensure the command and arguments match your installation and desired mode.

```json
{
  "mcpServers": {
    "conport": {
      "command": "uvx",
      "args": [
        "--from",
        "context-portal-mcp",
        "conport-mcp",
        "--mode",
        "stdio",
        "--workspace_id",
        "/path/to/your/project",
        "--log-file",
        "./logs/conport.log",
        "--log-level",
        "INFO"
      ]
    }
  }
}
```

--------------------------------

### Get System Patterns

Source: https://context7.com/greatscottymac/context-portal/llms.txt

Retrieves system patterns, with options to filter by tags. Useful for querying documented architectural or coding patterns.

```python
# Filter by tags
arch_patterns = await mcp.call_tool("get_system_patterns", {
    "workspace_id": "/path/to/my-project",
    "tags_filter_include_any": ["architecture", "design-pattern"],
    "limit": 10
})
```

```python
# Get all patterns
patterns = await mcp.call_tool("get_system_patterns", {
    "workspace_id": "/path/to/my-project"
})
```

--------------------------------

### Generic ConPort Strategy for LLMs

Source: https://github.com/greatscottymac/context-portal/blob/main/README.md

This platform-agnostic strategy file guides LLMs on using ConPort tools, emphasizing dynamic discovery of tool names and parameters via `get_conport_schema` for conceptual interactions.

```text
get_conport_schema
```

--------------------------------

### Get Item History using Python

Source: https://context7.com/greatscottymac/context-portal/llms.txt

Retrieve version history for Product Context or Active Context items using `get_item_history`. Supports fetching recent history, a specific version, or filtering by a timestamp range.

```python
# Get recent history
history = await mcp.call_tool("get_item_history", {
    "workspace_id": "/path/to/my-project",
    "item_type": "product_context",
    "limit": 5
})
```

```python
# Get a specific version
version_2 = await mcp.call_tool("get_item_history", {
    "workspace_id": "/path/to/my-project",
    "item_type": "product_context",
    "version": 2
})
```

```python
# Filter by timestamp range
recent_changes = await mcp.call_tool("get_item_history", {
    "workspace_id": "/path/to/my-project",
    "item_type": "active_context",
    "after_timestamp": "2025-01-14T00:00:00Z",
    "before_timestamp": "2025-01-16T00:00:00Z"
})
```

--------------------------------

### Create Virtual Environment

Source: https://github.com/greatscottymac/context-portal/blob/main/README.md

Creates a virtual environment using uv.

```bash
uv venv
```

--------------------------------

### Create and Activate Virtual Environment with uv

Source: https://github.com/greatscottymac/context-portal/blob/main/CONTRIBUTING.md

Create a virtual environment using uv and activate it. This isolates project dependencies. Activation commands differ for Linux/macOS and Windows.

```bash
uv venv
source .venv/bin/activate # Linux/macOS
# .venv\Scripts\activate.bat # Windows Command Prompt
# .venv\Scripts\Activate.ps1 # Windows PowerShell
```

--------------------------------

### Clone the Context Portal Repository

Source: https://github.com/greatscottymac/context-portal/blob/main/CONTRIBUTING.md

Clone the project repository to your local machine using Git. Navigate into the cloned directory.

```bash
git clone https://github.com/GreatScottyMac/context-portal.git
cd context-portal
```

--------------------------------

### Get ConPort Schema

Source: https://github.com/greatscottymac/context-portal/blob/main/conport_mcp_deep_dive.md

Arguments for retrieving the ConPort tool schema.

```APIDOC
## POST /get_conport_schema

### Description
Arguments for retrieving the ConPort tool schema.

### Method
POST

### Endpoint
/get_conport_schema

### Parameters
#### Request Body
- **workspace_id** (string) - Required - Identifier for the workspace (e.g., absolute path)

### Request Example
{
  "workspace_id": "/path/to/workspace"
}

### Response
#### Success Response (200)
- **schema** (object) - The ConPort tool schema.

#### Response Example
{
  "schema": {
    "item_types": {
      "document": {
        "fields": [
          {"name": "title", "type": "string"},
          {"name": "content", "type": "string"}
        ]
      }
    }
  }
}
```

--------------------------------

### Get Item History

Source: https://github.com/greatscottymac/context-portal/blob/main/conport_mcp_deep_dive.md

Arguments for retrieving history of a context item.

```APIDOC
## POST /get_item_history

### Description
Arguments for retrieving history of a context item.

### Method
POST

### Endpoint
/get_item_history

### Parameters
#### Request Body
- **workspace_id** (string) - Required - Identifier for the workspace (e.g., absolute path)
- **item_type** (string) - Required - Type of the item: 'product_context' or 'active_context'
- **limit** (integer) - Optional - Maximum number of history entries to return (most recent first)
- **before_timestamp** (string) - Optional - Return entries before this timestamp (format: date-time)
- **after_timestamp** (string) - Optional - Return entries after this timestamp (format: date-time)
- **version** (integer) - Optional - Return a specific version

### Request Example
{
  "workspace_id": "/path/to/workspace",
  "item_type": "product_context",
  "limit": 5,
  "after_timestamp": "2023-10-27T10:00:00Z"
}

### Response
#### Success Response (200)
- **history** (array of objects) - A list of history entries for the item.
  - **timestamp** (string) - The timestamp of the history entry.
  - **version** (integer) - The version of the item.
  - **change_description** (string) - Description of the change.

#### Response Example
{
  "history": [
    {
      "timestamp": "2023-10-27T11:30:00Z",
      "version": 2,
      "change_description": "Updated description."
    },
    {
      "timestamp": "2023-10-27T10:00:00Z",
      "version": 1,
      "change_description": "Initial creation."
    }
  ]
}
```

--------------------------------

### Run Context Portal Server in STDIO Mode

Source: https://github.com/greatscottymac/context-portal/blob/main/conport_mcp_deep_dive.md

Use this command to run the server in STDIO mode, suitable for local MCP clients like IDE extensions. The `workspace_id` can be automatically detected if `${workspaceFolder}` is passed literally.

```bash
python src/context_portal_mcp/main.py --mode stdio --workspace_id "/path/to/your/workspace"
```

```bash
conport --mode stdio --workspace_id "/path/to/your/workspace"
```

--------------------------------

### Get ConPort Schema

Source: https://context7.com/greatscottymac/context-portal/llms.txt

Retrieves the JSON schema for all available ConPort tools and their arguments.

```APIDOC
## get_conport_schema

### Description
Retrieves the JSON schema for all available ConPort tools and their arguments, useful for AI assistants to understand available capabilities.

### Method
`call_tool`

### Endpoint
`get_conport_schema`

### Parameters
#### Request Body
- **workspace_id** (string) - Required - The ID of the workspace.

### Request Example
```json
{
    "workspace_id": "/path/to/my-project"
}
```

### Response
#### Success Response (200)
- **(object)** - A JSON object where keys are tool names and values are their respective JSON schemas.

#### Response Example
```json
{
    "log_decision": {
        "title": "LogDecisionArgs",
        "type": "object",
        "properties": {
            "workspace_id": {"type": "string", "description": "..."},
            "summary": {"type": "string", "minLength": 1},
            "rationale": {"type": "string"},
            "tags": {"type": "array", "items": {"type": "string"}}
        },
        "required": ["workspace_id", "summary"]
    }
}
```
```

--------------------------------

### GET /custom_data

Source: https://context7.com/greatscottymac/context-portal/llms.txt

Retrieves custom data entries. Supports filtering by category and key.

```APIDOC
## GET /custom_data

### Description
Retrieves custom data entries with optional filtering by category and key.

### Method
GET

### Endpoint
/custom_data

### Parameters
#### Query Parameters
- **workspace_id** (string) - Required - The ID of the workspace to retrieve data from.
- **category** (string) - Optional - Filters the results to a specific category.
- **key** (string) - Optional - Filters the results to a specific key within a category.

### Response
#### Success Response (200)
- **id** (integer) - Unique identifier for the custom data entry.
- **timestamp** (string) - ISO 8601 timestamp of when the data was last updated.
- **category** (string) - The category of the custom data.
- **key** (string) - The key of the custom data entry.
- **value** (object) - The actual custom data value, which can be any JSON object.

#### Response Example
```json
[
    {
        "id": 1,
        "timestamp": "2025-01-15T10:00:00+00:00",
        "category": "ProjectGlossary",
        "key": "SKU",
        "value": {"definition": "Stock Keeping Unit...", ...}
    }
]
```
```

--------------------------------

### Create and Activate Virtual Environment with venv

Source: https://github.com/greatscottymac/context-portal/blob/main/CONTRIBUTING.md

Create a virtual environment using Python's built-in venv module and activate it. Activation commands differ for Linux/macOS and Windows.

```bash
python3 -m venv .venv # Or 'python -m venv .venv'
source .venv/bin/activate # Linux/macOS
# .venv\Scripts\activate.bat # Windows Command Prompt
# .venv\Scripts\Activate.ps1 # Windows PowerShell
```

--------------------------------

### Get Recent Activity Summary

Source: https://github.com/greatscottymac/context-portal/blob/main/conport_mcp_deep_dive.md

Arguments for retrieving a summary of recent ConPort activity.

```APIDOC
## POST /get_recent_activity_summary

### Description
Arguments for retrieving a summary of recent ConPort activity.

### Method
POST

### Endpoint
/get_recent_activity_summary

### Parameters
#### Request Body
- **workspace_id** (string) - Required - Identifier for the workspace (e.g., absolute path)
- **hours_ago** (integer) - Optional - Look back this many hours for recent activity. Mutually exclusive with 'since_timestamp'.
- **since_timestamp** (string) - Optional - Look back for activity since this specific timestamp (format: date-time). Mutually exclusive with 'hours_ago'.
- **limit_per_type** (integer) - Optional - Maximum number of recent items to show per activity type (e.g., 5 most recent decisions). (Default: 5)

### Request Example
{
  "workspace_id": "/path/to/workspace",
  "hours_ago": 24,
  "limit_per_type": 3
}

### Response
#### Success Response (200)
- **summary** (object) - A summary of recent activity.
  - **activity_type** (string) - The type of activity (e.g., 'created_item', 'updated_item').
  - **items** (array of objects) - List of recent items for this activity type.

#### Response Example
{
  "summary": {
    "created_item": [
      {
        "item_type": "decision",
        "item_id": "dec003",
        "timestamp": "2023-10-27T12:00:00Z"
      }
    ],
    "updated_item": [
      {
        "item_type": "document",
        "item_id": "doc123",
        "timestamp": "2023-10-27T11:30:00Z"
      }
    ]
  }
}
```

--------------------------------

### Get Workspace Detection Info

Source: https://context7.com/greatscottymac/context-portal/llms.txt

Provides diagnostic information about workspace detection for debugging and verification.

```APIDOC
## get_workspace_detection_info

### Description
Provides diagnostic information about workspace detection for debugging and verification.

### Method
`call_tool`

### Endpoint
`get_workspace_detection_info`

### Parameters
#### Request Body
- **start_path** (string) - Required - The starting path for workspace detection.

### Request Example
```json
{
    "start_path": "/path/to/project/src"
}
```

### Response
#### Success Response (200)
- **start_path** (string) - The path from which detection started.
- **detected_workspace** (string) - The path of the detected workspace.
- **detection_method** (string) - The method used for detection.
- **indicators_found** (array) - List of indicators found during detection.
- **server_version** (string) - The version of the server.
- **detection_timestamp** (string) - The timestamp of the detection.
- **auto_detection_available** (boolean) - Whether auto-detection is available.

#### Response Example
```json
{
    "start_path": "/path/to/project/src",
    "detected_workspace": "/path/to/project",
    "detection_method": "strong_indicators",
    "indicators_found": ["package.json", ".git"],
    "server_version": "0.3.13",
    "detection_timestamp": "2025-01-15T10:00:00.000000",
    "auto_detection_available": true
}
```
```

--------------------------------

### Context Budget Checklist

Source: https://github.com/greatscottymac/context-portal/blob/main/conport-custom-instructions/mem4sprint.md

Defines items to keep in AI context and items to fetch from ConPort to manage the context budget effectively.

```yaml
context_budget_checklist:
    keep_in_ai_context_only:
      - "active_context: { mode, sprint_id, focus }"
      - "Current goal/story/subtasks: IDs + 1-line summaries"
      - "Top blockers (<=3) with owners"
      - "Latest decisions (<=3) relevant to task"
      - "IDs/tags to fetch key docs (requirements, architecture, sprint_plan)"
    fetch_from_conport:
      - "All docs (artifact_kind: doc), RFCs, literature refs"
      - "Full decision log, progress history, metrics"
      - "Broader backlog stories beyond those active"
```

--------------------------------

### Get Decisions API

Source: https://context7.com/greatscottymac/context-portal/llms.txt

Retrieves logged decisions with optional filtering by tags and limit on results returned.

```APIDOC
## GET /get_decisions

### Description
Retrieves logged decisions with optional filtering by tags and limit on results returned.

### Method
GET

### Endpoint
/get_decisions

### Parameters
#### Query Parameters
- **workspace_id** (string) - Required - The ID of the workspace.
- **limit** (integer) - Optional - The maximum number of decisions to return.
- **tags_filter_include_all** (array of strings) - Optional - Filters decisions that have ALL specified tags.
- **tags_filter_include_any** (array of strings) - Optional - Filters decisions that have ANY of the specified tags.

### Request Example
```json
{
  "workspace_id": "/path/to/my-project",
  "limit": 10
}
```
```json
{
  "workspace_id": "/path/to/my-project",
  "tags_filter_include_all": ["authentication", "security"]
}
```
```json
{
  "workspace_id": "/path/to/my-project",
  "tags_filter_include_any": ["database", "api", "architecture"],
  "limit": 5
}
```

### Response
#### Success Response (200)
- **id** (integer) - The unique identifier for the decision.
- **timestamp** (string) - The timestamp when the decision was logged.
- **summary** (string) - The summary of the decision.
- **rationale** (string) - The rationale for the decision.
- **tags** (array of strings) - Tags associated with the decision.

#### Response Example
```json
[
    {
        "id": 3,
        "timestamp": "2025-01-15T14:20:00+00:00",
        "summary": "Use PostgreSQL with read replicas",
        "rationale": "Need high read throughput for product catalog...",
        "tags": ["database", "architecture"]
    },
    {
        "id": 2,
        "timestamp": "2025-01-15T12:00:00+00:00",
        "summary": "RESTful API design with OpenAPI spec",
        "rationale": "Standard REST patterns enable better tooling...",
        "tags": ["api", "architecture"]
    }
]
```
```

--------------------------------

### Get Subtasks of a Specific Parent

Source: https://context7.com/greatscottymac/context-portal/llms.txt

Retrieves a list of subtasks associated with a given parent task ID, with optional limits.

```APIDOC
## GET /get_progress

### Description
Retrieves subtasks of a specific parent task.

### Method
GET

### Endpoint
/get_progress

### Query Parameters
- **workspace_id** (string) - Required - The ID of the workspace.
- **parent_id_filter** (integer) - Required - The ID of the parent task to filter subtasks by.
- **limit** (integer) - Optional - The maximum number of subtasks to return.

### Response
#### Success Response (200)
- **id** (integer) - The unique identifier of the subtask.
- **timestamp** (string) - The timestamp when the subtask was created or last updated.
- **status** (string) - The current status of the subtask (e.g., DONE, IN_PROGRESS).
- **description** (string) - A description of the subtask.
- **parent_id** (integer) - The ID of the parent task.

#### Response Example
```json
[
    {
        "id": 2,
        "timestamp": "2025-01-15T11:00:00+00:00",
        "status": "DONE",
        "description": "Create JWT token generation service",
        "parent_id": 1
    },
    {
        "id": 3,
        "timestamp": "2025-01-15T11:30:00+00:00",
        "status": "IN_PROGRESS",
        "description": "Add token validation middleware",
        "parent_id": 1
    }
]
```
```

--------------------------------

### Get Project Subtasks

Source: https://context7.com/greatscottymac/context-portal/llms.txt

Retrieves subtasks associated with a specific parent task. Useful for tracking the progress of a larger initiative.

```python
subtasks = await mcp.call_tool("get_progress", {
    "workspace_id": "/path/to/my-project",
    "parent_id_filter": 1,
    "limit": 20
})
```

--------------------------------

### uvx Configuration for MCP Client

Source: https://github.com/greatscottymac/context-portal/blob/main/README.md

Recommended configuration for most IDEs using uvx to run the ConPort server. It includes arguments for source, destination, mode, workspace ID, and logging.

```json
{
  "mcpServers": {
    "conport": {
      "command": "uvx",
      "args": [
        "--from",
        "context-portal-mcp",
        "conport-mcp",
        "--mode",
        "stdio",
        "--workspace_id",
        "${workspaceFolder}",
        "--log-file",
        "./logs/conport.log",
        "--log-level",
        "INFO"
      ]
    }
  }
}
```

--------------------------------

### Get Progress API

Source: https://context7.com/greatscottymac/context-portal/llms.txt

Retrieves progress entries with optional filtering by status, parent task ID, and result limit.

```APIDOC
## GET /get_progress

### Description
Retrieves progress entries with optional filtering by status, parent task ID, and result limit.

### Method
GET

### Endpoint
/get_progress

### Parameters
#### Query Parameters
- **workspace_id** (string) - Required - The ID of the workspace.
- **status_filter** (string) - Optional - Filters progress entries by status (e.g., "IN_PROGRESS").
- **parent_id_filter** (integer) - Optional - Filters progress entries by parent task ID.
- **limit** (integer) - Optional - The maximum number of results to return.

### Request Example
```json
{
  "workspace_id": "/path/to/my-project",
  "status_filter": "IN_PROGRESS"
}
```
```

--------------------------------

### Dynamic Context Retrieval for RAG

Source: https://github.com/greatscottymac/context-portal/blob/main/conport-custom-instructions/mem4sprint.md

Outlines the steps for retrieving specific project knowledge for RAG. It includes analyzing the query, retrieving narrowly, handling sparse results, and synthesizing context.

```yaml
dynamic_context_retrieval_for_rag:
    trigger: "Need specific project knowledge to answer/generate."
    steps:
      - "Analyze query: entities/terms/item types needed."
      - "Retrieve narrowly: search_decisions_fts, search_custom_data_value_fts, search_project_glossary_fts."
      - "If specific items/IDs implied: get_custom_data/get_decisions/get_system_patterns/get_progress."
      - "Fallback if sparse: get_product_context or get_active_context (be brief)."
      - "Optionally 1-hop expand: get_linked_items for top candidates."
      - "Sift + synthesize concise context; attribute sources briefly."
    note: "Prefer semantic_search_conport for conceptual queries when keywords are weak. Keep retrieved context small and targeted."
```

--------------------------------

### Build Docker Image Manually

Source: https://github.com/greatscottymac/context-portal/blob/main/CONTRIBUTING.md

Manually build the Docker image using the 'docker build' command. Tag with 'latest' or a specific version.

```bash
docker build -t greatscottymac/context-portal-mcp:latest .

```

```bash
# docker build -t greatscottymac/context-portal-mcp:vX.Y.Z .

```

--------------------------------

### Get Linked Items

Source: https://github.com/greatscottymac/context-portal/blob/main/conport_mcp_deep_dive.md

Retrieves links connected to a specific ConPort item, allowing traversal of the project knowledge graph.

```APIDOC
## POST /get_linked_items

### Description
Arguments for retrieving links connected to a specific ConPort item, allowing traversal of the project knowledge graph.

### Method
POST

### Endpoint
/get_linked_items

### Parameters
#### Request Body
- **workspace_id** (string) - Required - Identifier for the workspace (e.g., absolute path)
- **item_type** (string) - Required - Type of the item to find links for (e.g., 'decision')
- **item_id** (string) - Required - ID or key of the item to find links for
- **relationship_type_filter** (string) - Optional - Filter by relationship type
- **linked_item_type_filter** (string) - Optional - Filter by the type of the linked items
- **limit** (integer) - Optional - Maximum number of links to return

### Request Example
{
  "workspace_id": "/path/to/workspace",
  "item_type": "document",
  "item_id": "doc123",
  "relationship_type_filter": "satisfies",
  "limit": 10
}

### Response
#### Success Response (200)
- **links** (array of objects) - A list of links connected to the item.
  - **relationship_type** (string) - The type of relationship.
  - **linked_item_type** (string) - The type of the linked item.
  - **linked_item_id** (string) - The ID of the linked item.
  - **description** (string) - Description of the link.

#### Response Example
{
  "links": [
    {
      "relationship_type": "satisfies",
      "linked_item_type": "requirement",
      "linked_item_id": "req456",
      "description": "This document satisfies the requirement."
    }
  ]
}
```

--------------------------------

### Build Docker Image using build.ps1

Source: https://github.com/greatscottymac/context-portal/blob/main/CONTRIBUTING.md

Use this PowerShell script for building the Docker image on Windows. It automates the build and tagging process.

```powershell
./build.ps1

```

--------------------------------

### Run Context Portal Server in HTTP Mode

Source: https://github.com/greatscottymac/context-portal/blob/main/conport_mcp_deep_dive.md

Execute this command to run the server as an HTTP service using Uvicorn. The MCP endpoint is typically available at `/mcp`.

```bash
uvicorn src.context_portal_mcp.main:app --host 127.0.0.1 --port 8000
```

```bash
conport --mode http --host 127.0.0.1 --port 8000
```

--------------------------------

### Retrieve Project Product Context

Source: https://context7.com/greatscottymac/context-portal/llms.txt

Use the `get_product_context` tool to retrieve high-level, static project information. Requires a `workspace_id`.

```python
# MCP Tool Call
result = await mcp.call_tool("get_product_context", {
    "workspace_id": "/path/to/my-project"
})

# Response
{
    "project_goal": "Build a scalable e-commerce platform",
    "key_features": [
        "User authentication",
        "Product catalog",
        "Shopping cart",
        "Payment processing"
    ],
    "architecture": "Microservices with event-driven communication",
    "tech_stack": {
        "frontend": "React",
        "backend": "Python FastAPI",
        "database": "PostgreSQL"
    }
}
```

--------------------------------

### Get Workspace Detection Info

Source: https://context7.com/greatscottymac/context-portal/llms.txt

Obtain diagnostic information about workspace detection. This tool is useful for debugging and verifying how a project workspace is identified.

```python
info = await mcp.call_tool("get_workspace_detection_info", {
    "start_path": "/path/to/project/src"
})
```

--------------------------------

### Get Recent Activity Summary

Source: https://context7.com/greatscottymac/context-portal/llms.txt

Retrieves a summary of recent activity within a workspace, either from the last 24 hours or since a specific timestamp.

```APIDOC
## Get Recent Activity Summary

### Description
Retrieves a summary of recent activity within a workspace. This can be filtered by a time range (last 24 hours) or a specific timestamp.

### Method
`call_tool`

### Endpoint
`get_recent_activity_summary`

### Parameters
#### Request Body
- **workspace_id** (string) - Required - The ID of the workspace.
- **hours_ago** (integer) - Optional - Number of hours to look back for activity.
- **since_timestamp** (string) - Optional - ISO 8601 timestamp to retrieve activity since.
- **limit_per_type** (integer) - Optional - Maximum number of items to return per activity type.

### Request Example (Last 24 hours)
```json
{
    "workspace_id": "/path/to/my-project",
    "hours_ago": 24,
    "limit_per_type": 5
}
```

### Request Example (Since Timestamp)
```json
{
    "workspace_id": "/path/to/my-project",
    "since_timestamp": "2025-01-14T00:00:00Z",
    "limit_per_type": 10
}
```

### Response
#### Success Response (200)
- **recent_decisions** (array) - List of recent decisions.
- **recent_progress** (array) - List of recent progress items.
- **recent_system_patterns** (array) - List of recent system patterns.
- **recent_custom_data** (array) - List of recent custom data entries.

#### Response Example
```json
{
    "recent_decisions": [
        {"id": 5, "summary": "Use GraphQL for mobile API", "timestamp": "..."}
    ],
    "recent_progress": [
        {"id": 12, "status": "DONE", "description": "Completed auth flow"}
    ],
    "recent_system_patterns": [],
    "recent_custom_data": [
        {"category": "api_endpoints", "key": "POST /graphql"}
    ]
}
```
```

--------------------------------

### Create RFC Document

Source: https://github.com/greatscottymac/context-portal/blob/main/conport-custom-instructions/mem4sprint.schema_and_templates.md

Logs a new RFC document. Ensure `<ABS_PATH>` is replaced with the workspace ID and `<rfc-id>` with the RFC's unique identifier. The `link_or_path` should point to the RFC file.

```javascript
mcp0_log_custom_data({ workspace_id: "<ABS_PATH>", category: "rfc_doc", key: "<rfc-id>", value: { title: "<title>", link_or_path: "mdc:/tasks/rfc/<file>.md", status: "draft", tags: ["rfc"], provenance: { agent: "architect", tool: "mem4sprint", ts: "" } } })
```

--------------------------------

### Find and Terminate Processes on Linux/macOS

Source: https://github.com/greatscottymac/context-portal/blob/main/v0.2.4_UPDATE_GUIDE.md

Use these commands in Bash or Zsh to find the context_portal_mcp.main process by its PID and terminate it.

```bash
ps aux | grep context_portal_mcp.main
kill <PID>
```