Claude Code Integration
Claude Code is Anthropic's official CLI for Claude with native MCP support. This guide covers multiple ways to connect Snipara to Claude Code.
Recommended Integration
Claude Code has the best MCP integration and is the preferred way to use Snipara. Native HTTP transport means zero local installation required.
What You Get
43 MCP Tools
Full access to context queries, memory, search, planning, and team features directly from Claude Code.
Zero Install
HTTP transport means no local packages needed. Just add configuration and you're ready.
Auto Tool Selection
Claude automatically uses Snipara tools when relevant to your questions about code and documentation.
Quick Setup (One Command)
The fastest way to connect Snipara to Claude Code is using the claude mcp add command:
claude mcp add my-project \ --transport http \ --url "https://api.snipara.com/mcp/PROJECT_SLUG" \ --header "X-API-Key: YOUR_API_KEY"Replace the placeholders:
my-project- A name for this MCP server (use your project slug)PROJECT_SLUG- Your project ID from the Snipara dashboardYOUR_API_KEY- Your API key from the API Keys tab
Manual Configuration
For more control, you can manually edit the Claude Code settings file at ~/.claude/settings.json:
{ "mcpServers": { "my-project": { "transport": "http", "url": "https://api.snipara.com/mcp/PROJECT_SLUG", "headers": { "X-API-Key": "YOUR_API_KEY" } } }}Verify Installation
After adding the server, verify it's configured correctly:
claude mcp listmy-project (http) - https://api.snipara.com/mcp/PROJECT_SLUGTest the connection:
claude mcp test my-project✓ Connection successful (45ms)✓ 43 tools availableUsing Snipara Tools
Once connected, Claude Code will automatically use Snipara tools when appropriate. You can also explicitly request them:
claudeUse rlm_ask to find how authentication works in this project[Calling rlm_ask with question="How does authentication work?"]Based on the documentation, authentication uses JWT tokens...Available MCP Tools
Snipara provides 43 MCP tools organized by category. Here are the most commonly used:
| Category | Tool | Description |
|---|---|---|
| Context Optimization | ||
| Primary | rlm_context_query | Full semantic search with token budgeting |
| Primary | rlm_ask | Quick query with predictable tokens (~2.5K) |
| Primary | rlm_search | Regex pattern search across docs |
| Agent Memory | ||
| Memory | rlm_remember | Store facts, decisions, learnings for later |
| Memory | rlm_recall | Semantically recall relevant memories |
| Memory | rlm_memories | List and browse stored memories |
| Planning & Multi-Query | ||
| Planning | rlm_plan | Generate execution plan for complex tasks |
| Planning | rlm_decompose | Break queries into sub-queries with deps |
| Planning | rlm_multi_query | Execute multiple queries in parallel |
| Session & Utility | ||
| Session | rlm_inject | Set persistent session context |
| Session | rlm_context | Show current session context |
| Utility | rlm_stats | Documentation statistics |
| Utility | rlm_sections | List all document sections |
Context Optimization
rlm_context_queryFull semantic search with token budgeting
rlm_askQuick query with predictable tokens (~2.5K)
rlm_searchRegex pattern search across docs
Agent Memory
rlm_rememberStore facts, decisions, learnings for later
rlm_recallSemantically recall relevant memories
rlm_memoriesList and browse stored memories
Planning & Multi-Query
rlm_planGenerate execution plan for complex tasks
rlm_decomposeBreak queries into sub-queries with deps
rlm_multi_queryExecute multiple queries in parallel
Session & Utility
rlm_injectSet persistent session context
rlm_contextShow current session context
rlm_statsDocumentation statistics
rlm_sectionsList all document sections
See the MCP Tools Reference for complete documentation of all 43 tools.
Multiple Projects
You can connect multiple Snipara projects to Claude Code:
claude mcp add frontend-docs --transport http --url ".../mcp/proj1" --header "..."claude mcp add backend-docs --transport http --url ".../mcp/proj2" --header "..."claude mcp add api-specs --transport http --url ".../mcp/proj3" --header "..."Claude Code will intelligently route questions to the appropriate project based on context.
Using Environment Variables
For security, store your API key in an environment variable:
~/.zshrc or ~/.bashrcexport SNIPARA_API_KEY="rlm_your_api_key_here"Then reference it in your configuration:
claude mcp add my-project \ --transport http \ --url "https://api.snipara.com/mcp/PROJECT_SLUG" \ --header "X-API-Key: $SNIPARA_API_KEY"Troubleshooting
- Run
claude mcp listto verify the server is added - Check the URL is correct and accessible
- Try
claude mcp remove my-projectand add it again
- Verify your API key is correct (no extra spaces)
- Check the header format:
X-API-Key: rlm_... - Ensure the API key hasn't expired
- Try generating a new API key in the dashboard
- Use explicit prompts: "Use rlm_ask to..."
- Check that documents are indexed in your project
- Verify tools with
claude mcp test my-project
- Check your network connection
- Large documentation sets may take longer to search
- Use more specific queries for faster results
Claude Code Plugin
For an enhanced experience, install the Snipara Claude Code Plugin. The plugin adds 14 slash commands, 5 auto-invoked skills, and automation hooks on top of the MCP server integration.
14 Slash Commands
/snipara:lite-mode, /snipara:remember, /snipara:plan, and more.
5 Auto-Invoked Skills
Claude auto-queries docs, recalls context, and plans complex tasks.
RLM Runtime
Safe code execution in Docker with trajectory logging.
/plugin marketplace add Snipara/snipara-claude/plugin install snipara@snipara-plugins