When an AI coding agent does something wrong, it's hard to know why — the agent's "thinking" happened once, live, and by the time you notice the bug that exact moment is gone. OrcaReplay solves this by recording everything the agent did like a flight recorder, so you can rewind and watch the run again offline for free, or even swap in a different AI model at any checkpoint to see if it would have handled the same situation better. OrcaReplay is an open-source (Apache-2.0) Node.js tool that captures a coding agent's full execution — proxy traffic, shell commands, MCP tool calls, filesystem changes — and lets you replay it byte-for-byte without hitting the network or paying for tokens again. It supports capture from Claude Code, Codex CLI, the OpenAI Agents SDK, Vercel AI SDK, grok-cli, LangGraph/LangChain and custom agents, and its standout feature is "model forking": replay up to a checkpoint, then continue with a different LLM to compare behavior on an identical prior context. It also builds causal graphs showing what caused what during a run, exportable as single-file HTML, SVG or PNG.
| # | Tool | Best for | Price |
|---|---|---|---|
| 1 | Développeurs | — | |
| 2 | Équipes d'ingénierie sur GitHub/GitLab voulant une génération de PR autonome et structurée plutôt qu'un usage ad-hoc d'agents IA | — | |
| 3 | Mainteneurs open source et équipes dev qui perdent du temps à reproduire manuellement des bugs signalés | — | |
| 4 | Développeurs construisant des agents de code auto-améliorants ayant besoin d'un historique d'exécution consultable | — | |
| 5 | Équipes SRE et platform engineering voulant une analyse de causes racines assistée par IA, avec approbation humaine obligatoire | — | |
| 6 | Équipes avec une infra conteneurisée/Kubernetes et une stack Grafana/Loki cherchant une maintenance de dépôt autonome | — | |
| 7 | Écrivains, romanciers et créateurs construisant des univers fictifs détaillés et des récits complexes | — | |
| 8 | Développeurs construisant des systèmes RAG sur des documents longs (recherche, transcripts, documentation technique) | — | |
| 9 | Roboticiens, ingénieurs IA embarquée, équipes construisant des systèmes autonomes pilotés par LLM | — | |
| 10 | Équipes entreprise (finance, conformité, santé, juridique), développeurs RAG cherchant une alternative aux bases vectorielles | — | |
| 11 | Développeurs et équipes déployant des agents IA en production sans vouloir gérer leur propre infrastructure | — | |
| 12 | Développeurs utilisant plusieurs agents de code IA (Claude Code, Cursor, Aider, Goose…) et voulant les faire collaborer | — |
An autonomous multi-agent software team that runs inside your GitHub/GitLab repo, turning Discussions into merged, reviewed PRs with no human intervention.
A bot that automatically reproduces GitHub bug reports: it reads the issue with an AI, tries the steps in a disposable Docker sandbox, and posts back exactly what happened.
Observability for self-improving AI agents that writes execution telemetry as plain readable files inside the repository itself, so agents can read their own history with normal file tools.
An open-source incident-analysis copilot: ask it in plain English why something broke, and it queries your observability stack, correlates deploys, and proposes a root cause with evidence, with human approval required before it acts.
A self-hosted, autonomous AI engineer that watches production logs, turns Linear tickets into pull requests, and manages PR review cycles inside isolated sandboxes.
A free, open-source writing app that acts as an AI thinking partner for novelists and worldbuilders, keeping full context of your story instead of just autocompleting sentences.
An open-source RAG library that organizes documents into a nested tree instead of flat chunks, so an AI assistant retrieves one relevant piece per branch instead of repeating itself.
An open-source runtime that lets you plug any major AI model into physical robot hardware, with built-in safety limits, fleet management, and chat-app remote control.
A document search engine for AI that reads and reasons through a document's structure like a human would, instead of chopping it into pieces and matching by similarity like traditional RAG.
A cloud platform for running AI agents 24/7 — schedule them, keep them always-on, or trigger them on demand — without managing your own servers.
A shared workspace where different AI coding agents — Claude Code, Cursor, Aider, Goose, and others — collaborate in the same thread, share a browser, and hand off tasks to each other.
