buptwz

buptwz / holmes-kb

Public

Knowledge-driven troubleshooting assistant — structured, git-native KB with AI-powered import and intelligent conflict resolution

19
0
85% credibility
Found May 29, 2026 at 25 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Holmes is an AI-powered troubleshooting assistant that helps engineering teams capture and share debugging knowledge. When engineers solve problems with Holmes, the solution is automatically saved to a shared knowledge base that improves over time. The knowledge base tracks how often each solution is used and automatically promotes reliable entries to higher confidence levels, while aging out entries that haven't been referenced recently. Entries are stored as simple text files in a shared folder, so teams can collaborate using standard tools and never lose access to their knowledge.

How It Works

1
🛠️ You install Holmes

You install the Holmes command-line tool on your computer. It takes just a moment.

2
📁 You connect your team's knowledge base

You point Holmes to a shared folder where your team keeps all their troubleshooting notes and documentation.

3
🤖 You start the AI assistant

You launch Holmes and describe a problem you're facing — maybe a server keeps timing out.

4
🔍 Your assistant searches the knowledge base first

Before doing anything else, Holmes looks through your team's shared knowledge to see if anyone has solved this before.

5
Holmes either finds a solution or helps you investigate
✅
Found in knowledge base

Holmes shows you a matching entry with proven steps to try

🔎
Nothing found

Holmes helps you investigate and run diagnostic commands

6
🎉 Your problem gets solved

You work through the issue with Holmes's guidance until everything is working again.

7
📝 Your solution gets automatically saved

The moment you say 'it's fixed,' Holmes quietly creates a new entry capturing everything you learned — symptoms, cause, and resolution.

🌱 Your team's knowledge grows stronger

Next time anyone hits a similar issue, Holmes finds your solution automatically. Over time, the knowledge base gets smarter and more reliable.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 25 to 19 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is holmes-kb?

Holmes-kb is a Python-based troubleshooting assistant that transforms debugging sessions into a structured, self-improving knowledge base. When you resolve an incident, the agent automatically extracts the troubleshooting arc and saves it for the next engineer who hits the same issue. The KB lives in git as plain Markdown files with YAML frontmatter, so your entire team can collaborate without lock-in.

Why is it gaining traction?

The hook is the closed loop: every resolved incident automatically feeds the knowledge base without manual effort. Entries gain confidence through real usage (draft to verified to proven), and stale knowledge decays automatically. The git-native approach means your KB is version-controlled, diffable, and mergeable — with five automatic conflict resolution scenarios that eliminate the need for coordination.

Who should use this?

Platform engineering teams managing complex infrastructure who find themselves solving the same incidents repeatedly. SREs who want institutional memory that survives personnel turnover. Any team using Claude Code or similar AI coding assistants that wants to capture troubleshooting knowledge without changing their existing workflow.

Verdict

Early but promising. At 19 stars, this is still finding its audience, and the two-repo architecture adds setup complexity. The 0.85% credibility score reflects that maturity — docs are solid but test coverage needs expansion. Worth experimenting with if your team struggles with knowledge transfer in on-call rotations, but plan to invest time in the initial setup.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.