cc-session-search
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@cc-session-search search for 'error handling' in last 3 days"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Claude Code Session Search MCP Server
An MCP (Model Context Protocol) server that provides tools for searching and analyzing Claude Code conversation history.
Features
List Projects: View all Claude Code projects with session counts
List Sessions: Browse sessions for specific projects
List Recent Sessions: Find recent conversations across all projects
Analyze Sessions: Extract and analyze messages with role filtering
Search Conversations: Search for specific terms with context windows and time ranges
Get Message Details: Retrieve full content for specific messages
Summarize Conversations: AI-powered summarization of daily conversations
Related MCP server: Claude Code History MCP Server
Installation
Install dependencies:
uv syncRun the server:
uv run python server.pyAdd to Claude Code MCP config (
~/.config/claude/mcp.json):
{
"servers": {
"cc-session-search": {
"command": ["uv", "run", "python", "server.py"],
"cwd": "/path/to/cc-session-search"
}
}
}Requirements
Standard Claude Code installation (searches
~/.claude/projects/)Python 3.13+
MCP 1.2.0+
Usage
The server provides the following tools:
list_projects()
Lists all Claude Code projects with session counts and recent activity.
list_sessions(project_name, days_back=7)
Lists sessions for a specific project within the specified time range.
list_recent_sessions(days_back=1, project_filter=None)
Lists recent sessions across all projects.
analyze_sessions(days_back=1, role_filter="both", include_tools=False, project_filter=None)
Extracts and analyzes messages from sessions with filtering options.
search_conversations(query, days_back=2, context_window=1, case_sensitive=False, project_filter=None)
Searches conversations for specific terms with context windows.
get_message_details(session_id, message_indices)
Retrieves full content for specific messages by session ID and indices.
Development
The server is built using the official MCP Python SDK with low-level Server class for maximum control.
Key features:
Efficient response handling with content truncation
Metadata-first approach to minimize token usage
Support for date ranges and filtering
Cross-project search capabilities
License
MIT
Available Tools
8 toolsanalyze_sessionsC
Extract and analyze messages from sessions with filtering
Inputs: days_back, role_filter, project_filter, include_tools.
get_message_detailsA
Get full content for specific messages by session ID and message indices
Inputs: session_id (required), message_indices (required).
list_projectsB
List all Claude Code projects with session counts and activity
Inputs: no named parameters.
list_recent_sessionsC
List recent sessions across all projects by date range
Inputs: days_back, project_filter.
list_sessionsC
List sessions for a specific project
Inputs: project_name (required), days_back.
search_conversationsB
Search conversations for specific terms with context windows, role filtering, and time ranges
Inputs: query (required), context_window, days_back, project_filter, +4 more.
summarize_daily_conversationsB
Generate intelligent summary of conversations for a specific date using headless Claude analysis
Inputs: date (required), style, project_filter.
summarize_time_rangeB
Generate intelligent summary of conversations for a specific time range using headless Claude analysis
Inputs: start_time (required), end_time (required), style, project_filter.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 8 tools
Most tools have distinct purposes, but 'summarize_daily_conversations' and 'summarize_time_range' are very similar, differing only in time scope. There is also potential overlap between 'analyze_sessions' and 'search_conversations', though descriptions help differentiate.
All tools use a consistent verb_noun pattern in snake_case, such as 'list_projects', 'search_conversations', and 'summarize_time_range'. No mixing of conventions or vague verbs.
With 8 tools, the server is well-scoped for searching and analyzing Claude Code sessions. Each tool serves a clear purpose without unnecessary redundancy.
The tool set covers listing, searching, analyzing, and summarizing sessions fairly well. However, it lacks tools for retrieving session metadata or project details beyond listing, which are minor gaps.
Maintenance
Related MCP Connectors
Augments MCP Server - A comprehensive framework documentation provider for Claude Code
Search your AI chat history (ChatGPT, Claude, Codex) from any MCP client. Remote, private, read-only
An MCP server that provides congressional transcripts
Related MCP Servers
- AlicenseBqualityDmaintenanceMCP server that lets you search your Claude Code conversation history to find past solutions, track file changes, and learn from previous work.8163 npm180MIT
- AlicenseAqualityDmaintenanceAn MCP server that enables users to retrieve, filter, and search through Claude Code conversation history stored in local projects. It provides tools for listing projects and sessions, paginating through message history, and searching across conversations with keyword filtering.421 npm11MIT
- AlicenseNot gradedqualityDmaintenanceAn MCP server that enables semantic and keyword search over Claude Code conversation history stored locally, using hybrid search, local embeddings, and time-decay scoring.12 npmMIT
- AlicenseNot gradedqualityDmaintenanceA local MCP server that indexes and searches your Claude Code conversation history with both keyword and semantic search, fully private and running locally.MIT