Agent Sessions, created by open-source developer Alex D. (jazzyalex), is a native macOS viewer and harness for indexing, searching, and inspecting interaction logs across CLI coding agents. Instead of running prompts directly, it functions as an aggregator that parses session history, tool execution records, and agent artifacts generated by various terminal assistants. In daily use, the application discovers session stores created by supported coding agents such as Claude Code, Codex hooks, Antigravity, and other CLI tools. It provides a searchable desktop interface with configurable preferences to filter sessions by agent kind, hide zero-message empty runs, and audit multi-turn conversation traces alongside tool invocations. Agent Sessions distinguishes itself by focusing on post-execution observability and format verification across disparate agent frameworks. It maintains continuous format-checking tooling and stage fixtures to handle evolving schema changes across CLI agent ecosystems. The project is open source and hosted on GitHub, with ongoing maintenance targeting macOS environments and support for over a dozen agent discovery paths.
Multi-Agent Session Discovery — Automatically locates and indexes session transcripts across diverse CLI coding tools including Claude Code, Codex, and Antigravity. - Unified Transcript Search — Provides a centralized search interface to query prompts, agent reasoning, and output across historical agent executions. - Zero-Message Filtering — Configurable preference rules to hide abandoned or zero-message agent sessions from the main interface view. - Format Drift Auditing — Built-in testing fixtures and validation scripts to ensure compatibility across evolving agent schema versions. - Native Swift UI — Responsive desktop interface tailored for fast navigation and transcript inspection on macOS.
Use cases
Use Case: Auditing Multi-Agent Coding Workflows Developers using multiple terminal agents (such as Claude Code and Codex) can review previous agent runs in a single place to compare tool usage, prompt responses, and task outcomes.
Use Case: Debugging Agent Tool Executions Engineers building custom CLI agents or agent hooks can inspect the exact payload, session structure, and tool call sequence recorded in local session files.
Use Case: Searching Past Agent Solutions Developers looking for code snippets or shell solutions generated during previous agent sessions can use the unified search to locate context without manual directory searching.
Who it is for
Agent Sessions is built for developers who regularly run multiple CLI-based AI coding agents on macOS and need a centralized tool to search, audit, and inspect their execution history. It is less suited for developers seeking an interactive agent orchestrator to run code generation directly in the terminal.
Install guide
Installation Clone the repository and build the macOS project using Swift / Xcode (standard build method for the Swift repository): bash git clone https://github.com/jazzyalex/agent-sessions.git cd agent-sessions swift build -c release
First Run Launch the application to automatically detect existing local agent session directories (such as ~/.claude or local hook folders), or verify supported agents with the internal check utility: bash # Verify supported agent discovery formats ./Scripts/steward_check --list-agents
Example Session Open the application to browse indexed agent logs, or use internal auditing scripts to check session integrity: bash # Search across local agent session formats python3 Scripts/search_audit.py
Plandex — Plandex is an open-source terminal-based AI coding agent designed to handle complex, multi-file engineering tasks in large codebases. Created by…
OpenCode — OpenCode is an open-source AI coding agent developed by Anomaly that operates across terminal environments, desktop applications, and IDE extensions.…
OpenHands — OpenHands (formerly OpenDevin) is an open-source autonomous AI software development agent and orchestration platform. It provides an extensible runtime environment…
Subtask — Subtask is an open-source CLI agent harness and orchestration tool developed by zippoxer in Go. It is built to coordinate…
EchoCoding — EchoCoding is an open-source agent harness and audio-augmented developer tooling layer created by launsion-boop. Designed to run alongside AI coding…
pi-reflect — pi-reflect is an open-source agent harness and self-improvement extension for pi coding agents, developed by jo-inc. It inspects agent session…