FreeVisit Website
Tracked since2026
The Bottom Line
- Price
- Free, no paid tier. Compare plans
Key facts
- Monitors multiple coding agents (Cursor, Claude Code, Codex) in one place, tracking their activity, rules, and usage.
- Local-first design keeps all conversation history and agent data on your device, ensuring privacy.
- Provides suggestions and analytics to improve agent setup, catch drift, and keep your codebase rulebook current.
Pros
- Local-first architecture ensures all data stays on your device, enhancing privacy and security.
- Integrates with multiple popular coding agents (Cursor, Claude Code, Codex) for unified management.
- Provides actionable analytics and suggestions to continuously improve agent setup and reduce drift.
Cons
- Some advanced features like auto-fixes and analytics are still in development.
- Limited to a desktop app; no web or mobile version mentioned.
What is Blume.codes?
Blume is a local-first desktop application that provides comprehensive visibility and control over coding agents such as Cursor, Claude Code, and Codex. It monitors agent activity, tracks hidden files, rules, skills, and hooks, and surfaces usage metrics and suggestions to improve agent setup and performance. All conversation history is stored locally on your device, ensuring privacy.
Blume helps developers see what each agent is doing, review changes before applying them, and catch drift early. It offers analytics on agent corrections, steering, and frustration levels, and proposes fixes that can be previewed and approved. The tool is designed to be a quiet sidecar that keeps your rulebook honest and your agents moving in the right direction.
Built by Peder Aaby and Olav Ljosland, Blume aims to make the management of coding agents effortless, providing a single source of truth for intent across your project. Upcoming features include auto-fixes, team conflict resolution, and a central domain model for team-wide consistency.
Key Features
- Agent Overview: See when agents finish, are working, or need approval.
- Hidden Files & Rules Tracking: View rules, skills, hooks, and other files that shape agent behavior.
- Usage Tracking: Monitor Claude Code and Codex plan limits and token burn before hitting limits.
- Auto-Fixes (soon): Detect mismatches between agent setups and chat instructions, with proposed fixes to approve.
- Analytics (soon): Track whether your agent setup is improving or degrading over time with metrics like corrections, steering, and frustration.
- Local Domain Model: A single source of truth for intent across your project, stored locally.
- Suggestions: Receive actionable suggestions to improve agent rules and skills based on observed patterns.
- Diff Review: Review exact diffs before applying, dismissing, or saving suggestions for later.
Pricing
Free
Blume.codes is completely free to use with no hidden costs.
Reviews
$99Free with your review
Write a reviewReview Blume.codes, get a free AI guide
Share your experience and we will send you Improve Your Thinking Patterns Using ChatGPT, free.
Best Blume.codes Alternatives
Top alternatives based on features, pricing, and user needs.
LangChainFreemium
Build LLM-powered applications
Amazon QFree
The most capable generative AI-powered assistant for accelerating software development and leveraging internal data.
PiecesFreemium
Long-term memory for your whole workstream, powered by AI to boost developer productivity.
OpenAI CodexFreemium
Delegate coding tasks in plain language, get pull requests
GitHub CopilotPaid
AI pair programmer that suggests code
DevinPaid
AI software engineer by Cognition Labs
Still deciding?
Most buyers shortlist 2 or 3 tools before committing. Pull a side-by-side comparison or browse the full alternatives shortlist below.
Explore More
Blume.codes FAQ
Which coding agents does Blume support?
Blume works with Cursor, Claude Code, Codex, and omp Pi. It monitors their activity, rules, skills, and hooks.
How does Blume ensure my conversation history remains private?
Blume stores all conversation history locally on your device. It is designed local-first, meaning no data is sent to external servers unless you explicitly choose to share it.
Can I review changes before applying suggestions from Blume?
Yes. Blume shows you the exact diff for each suggestion, and you can preview, apply, dismiss, or save the suggestion for later review.
What kind of analytics does Blume provide?
Blume provides analytics on agent corrections, steering, and frustration levels, shown as percentages and percentiles. You can view trends over a 30-day comparison to see if your agent setup is improving or degrading.
Does Blume support team collaboration?
A central domain model for team-wide intent and conflict resolution is on the roadmap. Currently, Blume is focused on individual use with a local domain model.
How does Blume track usage limits for different providers?
Blume tracks how much of each provider you have left across Claude Code, Codex, and Cursor, so you can see your remaining plan limits and token burn before hitting a wall.
Can I use Blume with multiple agents simultaneously?
Yes. Blume provides a unified view of all your agents, showing what each is doing, the rules and instructions guiding its work, and when it finishes or needs approval.
What are 'auto-fixes' and when will they be available?
Auto-fixes detect mismatches between your agent setups and your chat instructions. Blume proposes the fix, and you approve it. This feature is marked as 'soon' on the roadmap.
Source: blume.codes