Feedbacks

Feedbacks is an open-source visual feedback and session replay tool that captures screenshots, videos, and debugging evidence for teams and coding agents.

Feedbacks is a free, open-source, and self-hosted visual feedback and session replay platform designed for development teams and coding agents. It captures screenshots, videos, session replays, and debugging evidence to provide the exact context needed for fixing bugs. Users can pin elements, annotate screenshots, suggest text replacements, and discuss changes directly beside the evidence. The platform integrates with coding agents like Codex and Claude using MCP, allowing agents to read conversation threads, session events, and team context. It runs on a team server hosted on your own infrastructure, paired with a Chrome extension for capturing feedback. The project is licensed under Apache-2.0 and supports bringing your entire team together on one server.

Feedbacks at a glance

Best for: Software engineer
Runs on: Web, Chrome extension, Self-hosted
Pricing: 100% free, Apache-2.0 open source
Released: October 2026
Category: AI coding tool
Website: feedbacks.softinator.ai

Key features

Capture screenshots and videos
Record session replays
Pin exact elements and annotate screenshots
Suggest text replacements for specific phrases
Download complete thread bundles with diagnostics
Connect coding agents via MCP
Host on your own infrastructure

Use it for

Reporting software bugs with visual evidence
Sharing session replays with developers and AI agents
Annotating UI elements for design changes
Discussing project changes beside debugging data

Alternatives to Feedbacks

Crit — It keeps human feedback localized and precise for AI agent code iterations.
agent.reviews — Real tool reviews written directly by coding agents after task completion.
Replay — Explores apps autonomously and provides root cause analysis without writing tests.
OpenCode — Works with 75+ model providers, local models, and existing subscriptions
CodeRabbit — Most-installed AI PR reviewer, built to check code from coding agents
Eval Skills — Guides coding agents from first execution to validated graders and experiments.
Mindwalk — Visualize how coding agents explore and edit files on a 3D repository map.
jevgrep — Reduces coding agent costs by providing targeted, relevant repository context.
Agentation — It translates clicks into CSS selectors and source paths for agents.
Agent Skills — Encodes senior-engineer workflows and quality gates into 24 skills.
Spotnote — Sends exact file, line, and column data to coding agents from any UI element.
pstack — Improves AI coding agent outcomes with rigorous workflows.

All Feedbacks alternatives