Skip to main content
Glama
README.md
# bragdoc-mcp

An MCP server that helps you capture professional achievements using the STAR method. Claude interviews you — asking questions iteratively — until each achievement is captured with quantified impact and business outcomes. Once your achievements are logged you can generate period summaries and performance review self-assessments, all from Claude Desktop.

## Requirements

- Node.js 18+
- Claude Desktop (or any MCP-compatible client)

## Installation

No install required. Run directly via npx:

```bash
npx -y bragdoc-mcp
```

## Claude Desktop Configuration

Add to your Claude Desktop MCP config file:

**macOS:** `~/Library/Application Support/Claude/claude_desktop_config.json`
**Windows:** `%APPDATA%\Claude\claude_desktop_config.json`

```json
{
  "mcpServers": {
    "bragdoc": {
      "command": "npx",
      "args": ["-y", "bragdoc-mcp"]
    }
  }
}
```

Restart Claude Desktop after saving.

## Data Storage

All data is stored locally at `~/.bragdoc/`:

- `brag.json` — your achievements
- `competencies.json` — your role competency framework
- `goals.json` — your OKRs for the current review cycle

## First Use

Once connected, try these prompts in Claude Desktop:

**Capture an achievement:**
> Use the "Add Achievement" prompt — Claude will interview you using the STAR method until every field is captured with quantified impact.

**Weekly review:**
> Use the "Weekly Update" prompt — Claude checks what you accomplished and suggests new achievements to capture.

**Set up context (do this once):**
> Use "Set Up Competencies" to enter your organisation's competency framework.
> Use "Set Up Goals" to enter your OKRs for the current review cycle.

**Generate a summary or review:**
> Use "Period Summary" with a period like "Q1 2026" or "March 2026".
> Use "Performance Review" with your review period to generate a structured self-assessment.

## Available Tools

| Tool | Description |
|------|-------------|
| `add_achievement` | Add an achievement with STAR fields |
| `list_achievements` | List achievements with optional date/tag/keyword filters |
| `get_achievement` | Retrieve a single achievement by ID |
| `update_achievement` | Update any field of an existing achievement |
| `delete_achievement` | Delete an achievement |
| `get_competencies` | Retrieve your competency list |
| `set_competencies` | Save your competency list |
| `get_goals` | Retrieve your goals list |
| `set_goals` | Save your goals list |

## Resources

Three MCP resources are exposed for direct data access:

- `bragdoc://achievements`
- `bragdoc://competencies`
- `bragdoc://goals`

## Local Development

```bash
git clone https://github.com/YOUR_USERNAME/bragdoc-mcp
cd bragdoc-mcp
npm install
npm run build
npm test
```

For local Claude Desktop development, use an absolute path in the config:

```json
{
  "mcpServers": {
    "bragdoc": {
      "command": "node",
      "args": ["/absolute/path/to/bragdoc-mcp/dist/index.js"]
    }
  }
}
```

TDQS

A4/5.0

Scored across 9 tools

Disambiguation5/5

Each tool targets a distinct operation on a specific resource: achievements have full CRUD plus list, competencies and goals have get/set. No overlap in functionality.

Naming Consistency5/5

All tool names follow a consistent verb_noun pattern using snake_case (e.g., add_achievement, set_competencies). Predictable and clear.

Tool Count5/5

9 tools cover the brag document domain well: CRUD for achievements, get/set for competencies and goals. Not too many or too few.

Completeness5/5

Achievements have full lifecycle (create, read, update, delete, list with filters). Competencies and goals have get and replace-all set, which is sufficient for typical use. No obvious gaps.

Maintenance

ActivityInactive
ResponsivenessNo issues