What AI agents does CASS support?

CASS supports session history from 11+ agents, including Claude Code, Codex, Gemini CLI, Cursor, Aider, ChatGPT, Pi-Agent, and Factory.

Does CASS support semantic search?

Yes, CASS offers lexical, semantic (vector similarity), and hybrid search modes to find results based on either exact terms or conceptual meaning.

How does CASS handle large search results for LLMs?

CASS provides token budget management flags such as --max-tokens, --limit, and --fields minimal to ensure the output fits within an agent's context window.

Can I search sessions from multiple computers?

Yes, CASS includes a remote sources feature that allows you to sync and search session history from multiple machines via SSH.

How do I use CASS within an AI agent session?

To prevent blocking the terminal with an interactive TUI, always use the --robot or --json flags for machine-readable output.

CASS: Coding Agent Session Search

Name: CASS: Coding Agent Session Search
Author: lev-os

bylev-os

•

开发者工具

Indexes and searches local history across multiple AI coding agents using a machine-optimized interface.

CASS (Coding Agent Session Search) is a high-performance utility that unifies fragmented history from over 11 AI coding agents—including Claude Code, Cursor, Aider, and Gemini—into a single searchable index. It is specifically designed to bridge the gap between different AI tools, allowing developers and agents to retrieve past solutions, context, and code snippets through lexical, semantic, or hybrid search modes. By providing a dedicated 'Robot Mode' with machine-readable JSON output and advanced token budget management, CASS enables AI agents to autonomously query their own past interactions and solve problems more effectively.

主要功能

01Unified indexing for 11+ agents including Claude Code, Cursor, Aider, and Codex.

02Multi-machine support to sync and search session history across remote hosts via SSH.

03AI-optimized Robot Mode for structured JSON output and machine-readable data.

042 GitHub stars

05Token budget management with field truncation and soft limits for LLM context safety.

06Advanced search algorithms including BM25 lexical, vector semantic, and hybrid modes.

使用场景

01Retrieving previously implemented solutions to recurring bugs across different projects.

02Providing AI agents with historical context of past sessions to improve decision-making.

03Performing cross-agent knowledge transfers when switching between different coding tools.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add lev-os/agents cass

For use in Claude.ai and ChatGPT