@crypto512/jicon-mcp 0.6.1 → 0.7.1

package/README.md

@@ -17,45 +17,184 @@ npm install -g @crypto512/jicon-mcp
 
 ## What is Jicon?
 
-Jicon is an MCP server that enables AI assistants like Claude to interact with your Jira, Confluence, and Tempo instances. It provides 39 powerful tools for:
+Jicon is an MCP server that enables AI assistants like Claude to interact with your Jira, Confluence, and Tempo instances. It provides 52 powerful tools for:
 
 - 🎫 **Jira**: Issue management, searching, project operations, Agile boards, sprints
 - 📄 **Confluence**: Page management, content search, spaces, attachments, comments
 - ⏱️ **Tempo**: Time tracking, worklog management, accounts, and teams
+- 📦 **Buffer**: Content pagination for handling large responses
 
 ## Features at a Glance
 
-### Jira (15 Tools)
+### Jira (19 Tools)
 - Search issues with JQL
 - Create, update, and transition issues
-- Manage comments and links
+- Manage comments, links, and watchers
 - Access boards and sprints
-- View projects and metadata
+- View projects, issue types, and metadata
+- Track worklogs (individual and recursive totals)
 
-### Confluence (13 Tools)
+### Confluence (14 Tools)
 - Search content with CQL
-- Create and update pages
+- Create, update, and delete pages
 - Navigate page hierarchies
 - Manage attachments
 - Add and view comments
 
-### Tempo (11 Tools)
+### Tempo (12 Tools)
 - Log and manage work time
 - View and update worklogs
 - Access accounts and teams
-- Retrieve team worklogs
+- Retrieve team and epic worklogs
 - Get user time tracking info
 
+### Buffer (5 Tools)
+- Retrieve content chunks from buffered responses
+- List active buffers and their metadata
+- Clear buffers to free memory
+- **Search buffered content with grep-like patterns**
+- **Edit buffered content with find/replace**
+
+### Workload (2 Tools)
+- Convert time between units (seconds, hours, days)
+- Sum multiple workload values with mixed units
+
+## Content Buffering
+
+Tools that return large content (Confluence pages, Jira issues, Tempo worklogs) use an in-memory buffering system for efficient content pagination.
+
+### How It Works
+
+1. **Initial Call**: When you call a tool like `confluence_get_page`, it fetches the full content and stores it in a buffer
+2. **Chunked Response**: The response includes a `bufferId`, the first chunk of content (default 5000 chars), and pagination info
+3. **Retrieve More**: Use `buffer_get_chunk` with the `bufferId` to get subsequent chunks
+
+### Response Format
+
+Buffered tools return:
+- `bufferId`: Unique identifier for retrieving more content
+- `content`: The current chunk of content
+- `offset`: Starting position of this chunk
+- `limit`: Maximum characters returned
+- `totalSize`: Total size of buffered content
+- `hasMore`: Whether more content is available
+
+### Example Usage
+
+```typescript
+// 1. Get a Confluence page (returns first 5000 chars)
+{
+  "pageId": "123456"
+}
+// Response: { bufferId: "buf_xxx", content: "...", hasMore: true, totalSize: 15000 }
+
+// 2. Get next chunk
+{
+  "bufferId": "buf_xxx",
+  "offset": 5000,
+  "limit": 5000
+}
+// Response: { content: "...", hasMore: true }
+
+// 3. Continue until hasMore: false
+```
+
+### Buffered Tools
+
+- `confluence_get_page` - Returns page content with buffering
+- `confluence_get_page_by_title` - Returns page content with buffering
+- `jira_get_issue` - Returns issue data with buffering
+- `tempo_get_worklogs` - Returns worklog data with buffering
+- `tempo_get_team_worklogs` - Returns team worklog data with buffering
+
+### Buffer Management
+
+- **TTL**: Buffers expire after 10 minutes by default; edited buffers reset to 1 day TTL
+- **List Buffers**: Use `buffer_list` to see active buffers
+- **Clear Buffers**: Use `buffer_clear` to free memory
+
+### Searching Buffer Content
+
+Use `buffer_grep` to search within buffered content with regex patterns:
+
+```typescript
+// Search for "meeting" with 2 lines of context before and after
+{
+  "bufferId": "buf_xxx",
+  "pattern": "meeting",
+  "-C": 2
+}
+
+// Case-insensitive search with 3 lines after each match
+{
+  "bufferId": "buf_xxx",
+  "pattern": "API",
+  "-i": true,
+  "-A": 3
+}
+
+// Count occurrences only
+{
+  "bufferId": "buf_xxx",
+  "pattern": "TODO",
+  "output_mode": "count"
+}
+```
+
+**Parameters:**
+- `pattern`: Regex pattern to search for
+- `-A`: Lines to show after each match
+- `-B`: Lines to show before each match
+- `-C`: Lines to show before AND after (overrides -A/-B)
+- `-i`: Case insensitive search
+- `-n`: Show line numbers (default: true)
+- `output_mode`: "content" (default) or "count"
+- `multiline`: Enable multiline mode where `.` matches newlines
+- `head_limit`: Limit to first N matches
+- `offset`: Skip first N matches
+
+### Editing Buffer Content
+
+Use `buffer_edit` to perform find/replace on buffered content:
+
+```typescript
+// Replace single occurrence
+{
+  "bufferId": "buf_xxx",
+  "old_string": "Meeting at 2pm",
+  "new_string": "Meeting at 3pm"
+}
+
+// Replace all occurrences
+{
+  "bufferId": "buf_xxx",
+  "old_string": "TODO",
+  "new_string": "DONE",
+  "replace_all": true
+}
+```
+
+**Important:** `buffer_edit` modifies content in memory only. To persist changes:
+- For Confluence: call `confluence_update_page` with the modified content
+- For Jira: call `jira_update_issue` with the modified fields
+
+**Workflow Example:**
+1. `confluence_get_page` → get bufferId
+2. `buffer_grep` → find text to replace
+3. `buffer_edit` → replace text in memory
+4. `buffer_get_chunk` → get full modified content
+5. `confluence_update_page` → persist changes
+
 ## Configuration
 
-### Configuration with `.jicon.json` (Recommended)
+### Configuration with `.jicon.json`
 
 Jicon supports a flexible configuration hierarchy with permission control:
 
 **Configuration Priority:**
 1. `.jicon.json` (project root) - **Highest priority**
 2. `$HOME/.config/jicon/jicon.json` (user config) - Medium priority
-3. Environment variables - Lowest priority (legacy)
+3. Environment variables - Lowest priority
 
 **Create `.jicon.json` in your project root:**
 
@@ -70,52 +209,51 @@ Jicon supports a flexible configuration hierarchy with permission control:
     "url": "https://confluence.example.com",
     "username": "your-email@example.com",
     "token": "your-api-token"
-  }
-}
-```
-
-**Note:** This configuration uses the default `readonly` mode. To enable write operations, add:
-```json
-{
+  },
   "permissions": {
     "mode": "full"
   }
 }
 ```
 
+**Note:** Tempo uses the same Jira URL and credentials - no separate configuration needed.
+
 **⚠️ Important:** Add `.jicon.json` to your `.gitignore` to avoid committing credentials!
 
 ### Permission Modes
 
-Control which tools are allowed to execute. **As of v0.3.0, the default mode is `"readonly"`** to implement the principle of least privilege.
+Control which tools are allowed to execute:
 
-#### 1. **Read-Only Mode** (default)
+#### 1. **Full Access** (default)
 ```json
-{ "permissions": { "mode": "readonly" } }
+{ "permissions": { "mode": "full" } }
 ```
-Only read operations (26 tools: 10 Jira + 8 Confluence + 8 Tempo).
-This is the **recommended default** for security - users must explicitly opt-in to write access.
+All 52 tools are available.
 
-#### 2. **Full Access**
+#### 2. **Read-Only Mode**
 ```json
-{ "permissions": { "mode": "full" } }
+{ "permissions": { "mode": "readonly" } }
 ```
-All 39 tools are available (read + write operations).
+Only read operations (36 read tools + auto-included dependencies: 14 Jira + 9 Confluence + 9 Tempo + 4 Buffer read tools, plus workload utilities).
 
 #### 3. **Custom Mode** with Virtual Actions
 
 Simplify configuration with **virtual actions** that group related tools:
 
-**Virtual Actions Available:**
-- `jira_read` - All 10 Jira read tools
+**Virtual Actions Available (9 total):**
+- `jira_read` - All 14 Jira read tools
 - `jira_write` - All 5 Jira write tools
-- `jira_all` - All 15 Jira tools
-- `confluence_read` - All 8 Confluence read tools
+- `jira_all` - All 19 Jira tools
+- `confluence_read` - All 9 Confluence read tools
 - `confluence_write` - All 5 Confluence write tools
-- `confluence_all` - All 13 Confluence tools
-- `tempo_read` - All 8 Tempo read tools
+- `confluence_all` - All 14 Confluence tools
+- `tempo_read` - All 9 Tempo read tools
 - `tempo_write` - All 3 Tempo write tools
-- `tempo_all` - All 11 Tempo tools
+- `tempo_all` - All 12 Tempo tools
+
+**Auto-included Dependencies:**
+- Buffer tools (buffer_get_chunk, buffer_grep, etc.) are automatically included with any read/write action
+- Workload tools (workload_convert, workload_sum) are automatically included with any Jira/Confluence/Tempo tool
 
 **Example: Read-only access to all services**
 ```json
@@ -166,11 +304,24 @@ Simplify configuration with **virtual actions** that group related tools:
 
 See [`.jicon.json.example`](.jicon.json.example), [`.jicon.json.readonly`](.jicon.json.readonly), and [`.jicon.json.custom`](.jicon.json.custom) for more examples.
 
-### Environment Variables (Legacy)
+### Environment Variables
 
-For Claude Desktop, add to config file:
+All configuration options can be set via environment variables:
 
-**MacOS**: `~/Library/Application Support/Claude/claude_desktop_config.json`
+| Variable | Description |
+|----------|-------------|
+| `JIRA_URL` | Jira instance URL |
+| `JIRA_USERNAME` | Username/email (omit for Bearer auth) |
+| `JIRA_API_TOKEN` | API token or PAT |
+| `JIRA_AUTH_TYPE` | `basic` (default) or `bearer` |
+| `CONFLUENCE_URL` | Confluence instance URL |
+| `CONFLUENCE_USERNAME` | Username/email (omit for Bearer auth) |
+| `CONFLUENCE_API_TOKEN` | API token or PAT |
+| `CONFLUENCE_AUTH_TYPE` | `basic` (default) or `bearer` |
+
+### Claude Desktop
+
+Add to `claude_desktop_config.json`:
 
 ```json
 {
@@ -191,6 +342,50 @@ For Claude Desktop, add to config file:
 }
 ```
 
+### Claude Code
+
+Add to `~/.claude/settings.json` or project `.mcp.json`:
+
+```json
+{
+  "mcpServers": {
+    "jicon": {
+      "command": "npx",
+      "args": ["-y", "@crypto512/jicon-mcp"],
+      "env": {
+        "JIRA_URL": "https://jira.example.com",
+        "JIRA_USERNAME": "your-email@example.com",
+        "JIRA_API_TOKEN": "your-api-token"
+      }
+    }
+  }
+}
+```
+
+### OpenCode
+
+For [OpenCode](https://opencode.ai) 1.0+, add to `opencode.jsonc` in your project root or `~/.config/opencode/opencode.jsonc`:
+
+```jsonc
+{
+  "$schema": "https://opencode.ai/config.json",
+  "mcp": {
+    "jicon": {
+      "type": "local",
+      "command": ["npx", "-y", "@crypto512/jicon-mcp"],
+      "environment": {
+        "JIRA_URL": "https://jira.example.com",
+        "JIRA_USERNAME": "your-email@example.com",
+        "JIRA_API_TOKEN": "your-api-token",
+        "CONFLUENCE_URL": "https://confluence.example.com",
+        "CONFLUENCE_USERNAME": "your-email@example.com",
+        "CONFLUENCE_API_TOKEN": "your-api-token"
+      }
+    }
+  }
+}
+```
+
 ## Example Usage
 
 Once configured, you can ask Claude to:
@@ -209,16 +404,19 @@ Once configured, you can ask Claude to:
 ### Jira Tools
 | Category | Tools |
 |----------|-------|
+| **Help** | `jira_jql_help` |
 | **Search & Retrieval** | `jira_search_issues`, `jira_get_issue`, `jira_get_issue_comments` |
 | **Issue Management** | `jira_create_issue`, `jira_update_issue`, `jira_transition_issue` |
 | **Collaboration** | `jira_add_comment`, `jira_link_issues`, `jira_get_issue_watchers` |
-| **Projects** | `jira_list_projects`, `jira_get_project` |
+| **Projects** | `jira_list_projects`, `jira_get_project`, `jira_get_issue_types` |
 | **Agile** | `jira_get_board`, `jira_get_sprints`, `jira_get_sprint_issues` |
 | **Workflow** | `jira_get_transitions` |
+| **Time Tracking** | `jira_get_issue_worklogs`, `jira_get_total_worklogs` |
 
 ### Confluence Tools
 | Category | Tools |
 |----------|-------|
+| **Help** | `confluence_cql_help` |
 | **Search & Retrieval** | `confluence_search_content`, `confluence_get_page`, `confluence_get_page_by_title` |
 | **Page Management** | `confluence_create_page`, `confluence_update_page`, `confluence_delete_page` |
 | **Spaces** | `confluence_list_spaces`, `confluence_get_space` |
@@ -231,9 +429,22 @@ Once configured, you can ask Claude to:
 |----------|-------|
 | **Worklog Management** | `tempo_get_worklogs`, `tempo_get_worklog`, `tempo_log_work`, `tempo_update_worklog`, `tempo_delete_worklog` |
 | **Team Tracking** | `tempo_get_team_worklogs`, `tempo_get_teams`, `tempo_get_team` |
+| **Epic Tracking** | `tempo_get_epic_worklogs` |
 | **Accounts** | `tempo_get_accounts`, `tempo_get_account` |
 | **User Info** | `tempo_get_user_info` |
 
+### Buffer Tools
+| Category | Tools |
+|----------|-------|
+| **Content Pagination** | `buffer_get_chunk`, `buffer_list`, `buffer_clear` |
+| **Content Search** | `buffer_grep` |
+| **Content Edit** | `buffer_edit` |
+
+### Workload Tools
+| Category | Tools |
+|----------|-------|
+| **Time Conversion** | `workload_convert`, `workload_sum` |
+
 ## Requirements
 
 - Node.js 18 or higher
@@ -244,26 +455,37 @@ Once configured, you can ask Claude to:
 
 ## Authentication
 
-### Jira/Confluence Cloud
+### Jira/Confluence Cloud (Basic Auth)
 Generate an API token from your Atlassian account:
 1. Go to https://id.atlassian.com/manage-profile/security/api-tokens
 2. Create a new API token
-3. Use your email as username and the token as password
+3. Use your email as `username` and the token as `token`
 
-### Jira/Confluence Data Center
-Generate a Personal Access Token (PAT):
+### Jira/Confluence Data Center (Bearer Auth)
+For Data Center 8.14+, use Personal Access Tokens (PAT):
 1. Go to your profile settings
 2. Navigate to "Personal Access Tokens"
 3. Create a new token with required permissions
+4. Omit `username` or set `authType: "bearer"` to use Bearer authentication
+
+**Data Center `.jicon.json` example:**
+```json
+{
+  "jira": {
+    "url": "https://jira.example.com",
+    "token": "your-personal-access-token",
+    "authType": "bearer"
+  }
+}
+```
 
 ## Security Best Practices
 
-- ✅ **Default Read-Only Mode**: Jicon defaults to `readonly` mode - write access must be explicitly enabled
-- ✅ Store credentials in `.jicon.json` and add it to `.gitignore`
+- ✅ Store credentials in environment variables
 - ✅ Use API tokens with minimal required permissions
 - ✅ Regularly rotate API tokens
 - ✅ Never commit credentials to version control
-- ✅ Start with `readonly` mode and only enable write operations when needed
+- ✅ Use read-only tokens for search/read operations
 
 ## License
 

package/TOOL_LIST.md

@@ -5,12 +5,13 @@ This document provides a comprehensive reference of all available tools in the J
 
 ## Summary
 
-**Total Tools**: 39
+**Total Tools**: 42
 - **Jira Tools**: 15
 - **Confluence Tools**: 13
 - **Tempo Tools**: 11
+- **Buffer Tools**: 3
 
-**Note**: Detailed documentation for Tempo tools is available in the source code at `src/tempo/tools.ts`. Tempo provides time tracking, worklog management, team tracking, and account management capabilities.
+**Note**: Tools that return large content use in-memory buffering with pagination support. See the Buffer Tools section for content retrieval.
 
 ---
 
@@ -29,7 +30,7 @@ This document provides a comprehensive reference of all available tools in the J
 ---
 
 ### 2. jira_get_issue
-**Description**: Get detailed information about a specific issue
+**Description**: Get detailed information about a specific issue with buffered content pagination
 **Use Cases**: View issue details, check status, read description and comments
 
 | Parameter | Type | Required | Description |
@@ -37,6 +38,10 @@ This document provides a comprehensive reference of all available tools in the J
 | issueKey | string | ✓ | Issue key (e.g., "PROJ-123") |
 | fields | string[] | ✗ | Specific fields to return |
 | expand | string[] | ✗ | Additional data (e.g., "changelog") |
+| offset | number | ✗ | Character offset to start from (default: 0) |
+| limit | number | ✗ | Maximum characters to return (default: 5000) |
+
+**Response includes**: `bufferId`, `content`, `offset`, `limit`, `totalSize`, `hasMore`
 
 ---
 
@@ -209,18 +214,22 @@ This document provides a comprehensive reference of all available tools in the J
 ---
 
 ### 2. confluence_get_page
-**Description**: Get a Confluence page by ID
+**Description**: Get a Confluence page by ID with buffered content pagination
 **Use Cases**: Read page content, check version
 
 | Parameter | Type | Required | Description |
 |-----------|------|----------|-------------|
 | pageId | string | ✓ | Page ID |
 | expand | string[] | ✗ | Additional data (e.g., "body.storage") |
+| offset | number | ✗ | Character offset to start from (default: 0) |
+| limit | number | ✗ | Maximum characters to return (default: 5000) |
+
+**Response includes**: `bufferId`, `content`, `offset`, `limit`, `totalSize`, `hasMore`, page metadata
 
 ---
 
 ### 3. confluence_get_page_by_title
-**Description**: Get a page by title and space
+**Description**: Get a page by title and space with buffered content pagination
 **Use Cases**: Find page by name, access documentation
 
 | Parameter | Type | Required | Description |
@@ -228,6 +237,10 @@ This document provides a comprehensive reference of all available tools in the J
 | spaceKey | string | ✓ | Space key |
 | title | string | ✓ | Page title |
 | expand | string[] | ✗ | Additional data to expand |
+| offset | number | ✗ | Character offset to start from (default: 0) |
+| limit | number | ✗ | Maximum characters to return (default: 5000) |
+
+**Response includes**: `bufferId`, `content`, `offset`, `limit`, `totalSize`, `hasMore`, page metadata
 
 ---
 
@@ -436,3 +449,186 @@ AND lastModified >= now("-30d")
 - **Meeting notes**: Recording decisions, action items
 - **Collaboration**: Comments, feedback on pages
 - **Content organization**: Space and page hierarchy management
+
+---
+
+## Tempo Tools (11)
+
+### 1. tempo_get_worklogs
+**Description**: Get time tracking worklogs with buffered content pagination
+**Use Cases**: View time entries, generate reports, review team work
+
+| Parameter | Type | Required | Description |
+|-----------|------|----------|-------------|
+| dateFrom | string | ✗ | Start date (YYYY-MM-DD) |
+| dateTo | string | ✗ | End date (YYYY-MM-DD) |
+| projectKey | string | ✗ | Filter by project key |
+| issueKey | string | ✗ | Filter by issue key |
+| workerKey | string | ✗ | Filter by worker username |
+| offset | number | ✗ | Character offset to start from (default: 0) |
+| limit | number | ✗ | Maximum characters to return (default: 5000) |
+
+**Response includes**: `bufferId`, `content`, `offset`, `limit`, `totalSize`, `hasMore`, worklog metadata
+
+---
+
+### 2. tempo_get_worklog
+**Description**: Get detailed information about a specific worklog entry
+**Use Cases**: View worklog details, check time logged
+
+| Parameter | Type | Required | Description |
+|-----------|------|----------|-------------|
+| worklogId | number | ✓ | Tempo worklog ID |
+
+---
+
+### 3. tempo_log_work
+**Description**: Log time to a Jira issue
+**Use Cases**: Record work time, track daily activities
+
+| Parameter | Type | Required | Description |
+|-----------|------|----------|-------------|
+| issueKey | string | ✓ | Jira issue key (e.g., PROJ-123) |
+| hours | number | ✓ | Hours to log (decimal, e.g., 2.5 = 2h 30m) |
+| date | string | ✓ | Date to log work (YYYY-MM-DD) |
+| description | string | ✗ | Work description |
+| startTime | string | ✗ | Start time (HH:MM:SS) |
+| remainingEstimateHours | number | ✗ | Remaining estimate in hours |
+
+---
+
+### 4. tempo_update_worklog
+**Description**: Update an existing worklog entry
+**Use Cases**: Correct time logged, update descriptions
+
+| Parameter | Type | Required | Description |
+|-----------|------|----------|-------------|
+| worklogId | number | ✓ | Tempo worklog ID to update |
+| hours | number | ✗ | New hours (decimal) |
+| description | string | ✗ | New description |
+| date | string | ✗ | New date (YYYY-MM-DD) |
+
+---
+
+### 5. tempo_delete_worklog
+**Description**: Delete a worklog entry
+**Use Cases**: Remove incorrect entries
+
+| Parameter | Type | Required | Description |
+|-----------|------|----------|-------------|
+| worklogId | number | ✓ | Tempo worklog ID to delete |
+
+---
+
+### 6. tempo_get_accounts
+**Description**: Get all Tempo accounts
+**Use Cases**: List accounts for time tracking, find account keys
+
+| Parameter | Type | Required | Description |
+|-----------|------|----------|-------------|
+| (none) | | | |
+
+---
+
+### 7. tempo_get_account
+**Description**: Get detailed information about a Tempo account
+**Use Cases**: View account details, check status
+
+| Parameter | Type | Required | Description |
+|-----------|------|----------|-------------|
+| accountKey | string | ✓ | Account key |
+
+---
+
+### 8. tempo_get_teams
+**Description**: Get all Tempo teams
+**Use Cases**: List teams, find team IDs
+
+| Parameter | Type | Required | Description |
+|-----------|------|----------|-------------|
+| (none) | | | |
+
+---
+
+### 9. tempo_get_team
+**Description**: Get detailed information about a Tempo team
+**Use Cases**: View team details, check membership
+
+| Parameter | Type | Required | Description |
+|-----------|------|----------|-------------|
+| teamId | number | ✓ | Team ID |
+
+---
+
+### 10. tempo_get_team_worklogs
+**Description**: Get all worklogs for a team with buffered content pagination
+**Use Cases**: Team capacity reports, sprint reviews
+
+| Parameter | Type | Required | Description |
+|-----------|------|----------|-------------|
+| teamId | number | ✓ | Team ID |
+| dateFrom | string | ✓ | Start date (YYYY-MM-DD) |
+| dateTo | string | ✓ | End date (YYYY-MM-DD) |
+| offset | number | ✗ | Character offset to start from (default: 0) |
+| limit | number | ✗ | Maximum characters to return (default: 5000) |
+
+**Response includes**: `bufferId`, `content`, `offset`, `limit`, `totalSize`, `hasMore`, worklog metadata
+
+---
+
+### 11. tempo_get_user_info
+**Description**: Get current user information (workerKey, accountId, displayName)
+**Use Cases**: Get workerKey for filtering worklogs, verify authentication
+
+| Parameter | Type | Required | Description |
+|-----------|------|----------|-------------|
+| (none) | | | |
+
+---
+
+## Buffer Tools (3)
+
+Buffer tools are used to retrieve content from buffered responses. When tools like `confluence_get_page` or `jira_get_issue` return large content, they store the full data in an in-memory buffer and return a `bufferId` for pagination.
+
+### 1. buffer_get_chunk
+**Description**: Retrieve a chunk of buffered content by buffer ID
+**Use Cases**: Get remaining content from large responses
+
+| Parameter | Type | Required | Description |
+|-----------|------|----------|-------------|
+| bufferId | string | ✓ | Buffer ID from previous tool response |
+| offset | number | ✗ | Character position to start from (default: 0) |
+| limit | number | ✗ | Maximum characters to return (default: 5000) |
+
+**Response includes**: `bufferId`, `content`, `offset`, `limit`, `totalSize`, `hasMore`, `metadata`
+
+**Example workflow**:
+```typescript
+// 1. Call confluence_get_page -> returns bufferId, first chunk, hasMore: true
+// 2. Call buffer_get_chunk(bufferId, offset: 5000) -> returns next chunk
+// 3. Repeat until hasMore: false
+```
+
+---
+
+### 2. buffer_list
+**Description**: List all active buffers with their metadata
+**Use Cases**: View active buffers, check expiration times
+
+| Parameter | Type | Required | Description |
+|-----------|------|----------|-------------|
+| (none) | | | |
+
+**Response includes**: `count`, list of buffers with `bufferId`, `totalSize`, `createdAt`, `expiresAt`, `metadata`
+
+---
+
+### 3. buffer_clear
+**Description**: Clear a specific buffer or all buffers
+**Use Cases**: Free up memory, clean up after processing
+
+| Parameter | Type | Required | Description |
+|-----------|------|----------|-------------|
+| bufferId | string | ✗ | Buffer ID to clear, or omit to clear all buffers |
+
+**Response includes**: `message`, `cleared` (count of buffers cleared)