Confessor: Analyzes Claude Code transcripts to show what the AI agent did on your computer
Confessor is an open-source tool that reads the JSONL log files Claude Code leaves on disk and generates a human-readable report of every action the AI agent performed, including tool calls, file reads, and shell commands. It helps developers understand and audit the behavior of AI coding agents on their machines.
Visit project ↗Entities
Related
Snitch: deterministic claim verifier for AI agent transcripts
Snitch is a tool that watches AI agent transcript files (Cursor, Claude Code, Codex, Pi, OpenCode) and verifies claims made in prose against actual evidence like tool calls, shell output, filesystem changes, git history, and session context. It uses deterministic regex patterns to extract claims and cross-references them, flagging inconsistencies. It helps developers trust their coding agents by catching when an agent's description doesn't match reality.
Ditto: mine your own LLM coding sessions into a file for your agent
Ditto extracts only the user's typed messages from local Claude Code and Codex session logs, stripping tool output and assistant replies, and compiles them into a file that an AI agent can read first. It solves the problem of losing personal work patterns and context across many coding sessions, giving the agent a honest record of how the user actually works.
Terminal dashboard for Claude Code session management
DejaView is a terminal UI that reads Claude Code transcripts from ~/.claude/projects/, groups sessions by project, and lets users resume any session with Enter. It solves the problem of remembering which directory a Claude session was in for the claude --resume command. Built with Textual, it is read-only, local-only, and requires no configuration.
Git-aware AI debugger that checks out old commits to fix production bugs
A tool that makes AI coding assistants (like Cursor or Claude Code) automatically checkout the git commit corresponding to a production error before debugging, preventing the agent from analyzing current code that has shifted. It solves the problem of AI agents hallucinating fixes because they look at the present state of files while the bug existed in a past commit.
Local CLI tool to analyze Claude AI token usage and costs
A Go-based CLI that parses local Claude AI logs to show token usage and cost breakdown by project, session, and model. It helps users understand where their API quota is going and how much they would have spent without caching.