Blume doesn't run agents itself - it watches your Claude Code, Cursor, and Codex conversations for repeated friction, then proposes the specific rule, skill, or hook that would have prevented it.
Badges
Badges come from GitHub once the author claims this feed.
Recent posts
every link opens the author’s post-
Blume: The Sidecar That Watches Your Coding Agents and Fixes Their Rulebook
-
How to Stop Spam Calls on iPhone: Screen, Silence and Filter Unknown Callers
Sick of spam calls on your iPhone?
-
Mac System Data Taking Up Huge Space? Here’s How to Shrink It
Mac System Data eating 100GB or more?
-
Laravel Pulse: Real-Time Application Performance Monitoring
A complete guide to Laravel Pulse - the production-facing performance monitoring dashboard covering slow routes, slow queries, exception rates, cache hit ratios, custom recorders, and async ingestion setup.
-
Claude Code Command Not Found? How to Fix Your PATH
Installed Claude Code but the terminal says claude: command not found?
-
Holm Security: Unified Vulnerability Management for IT, Cloud, and People
Holm Security's Next-Gen VM platform covers network, cloud, web application, and API scanning alongside phishing and awareness testing in a single European-built product.
-
Termux Permission Denied on Storage? How to Fix termux-setup-storage
Getting 'Permission denied' when Termux touches your phone's storage?
-
Persistent Memory for Claude: Five Approaches Compared
CLAUDE.md, claude-mem, Reporecall, the Private Journal MCP and MemSoph all give Claude a memory, but they remember very different things.
-
JetBrains AI Assistant with Local Models via Ollama and LM Studio
How to connect JetBrains AI Assistant to Ollama or LM Studio, assign local models to chat and code completion, and what local models can and cannot do inside the IDE.
-
Portainer vs Dockge: Managing Docker Without Living in Compose Files
Two ways to put a web UI on a Docker host, compared on the thing that matters most: whether your compose files stay the source of truth or quietly drift out of date.
-
iPhone Mirroring Not Working on Mac? Here’s How to Fix It
iPhone Mirroring not working on your Mac?
-
MemSoph: An AI Chat App With Memory You Control
MemSoph is an AI chat app that keeps the corrections you choose across sessions, with memory you can edit or delete.
-
ffuf vs Gobuster: Web Directory Brute Forcing on Kali Linux
ffuf and Gobuster both brute force hidden directories and files on Kali Linux.
-
Cloudflare’s security-audit Skill: A Six-Phase AI Vulnerability Hunt for Your Codebase
Cloudflare has open-sourced the security-audit skill that seeded its internal vulnerability discovery harness.
-
Is Claude Code Safe? The Real Risks and How to Lock It Down
Is Claude Code safe to run on your machine?
-
Nucleus Security: The Vulnerability Intelligence and Aggregation Layer
Nucleus Security doesn't scan anything itself - it aggregates and deduplicates findings from every scanner you already run, then manages the remediation workflow.
-
Fix ‘Could Not Get Lock /var/lib/dpkg/lock-frontend’ on Ubuntu and Kali
apt says it could not get lock /var/lib/dpkg/lock-frontend?
-
SSH in Termux: sshd, Keys and Connecting From Your Laptop
Run an SSH server in Termux so you can work on your phone from a proper keyboard, set up key-based login, and use Termux as an SSH client for your own servers.
-
Kimi K3: What Moonshot AI’s Giant Open-Weight Model Changes
Kimi K3 is Moonshot AI's 2.8-trillion-parameter open-weight model - here's how its Mixture-of-Experts architecture and 1M-token context actually compare to Qwen3.8 and GLM-5.3.
-
iPhone Battery Draining Fast After an iOS Update? How to Fix It
Some battery drain after an iOS update is normal while your iPhone indexes and downloads new assets.
-
Microsoft Defender Vulnerability Management: Built-In Scanning for the Defender Stack
Microsoft Defender Vulnerability Management extends the Defender for Endpoint agent into full vulnerability scanning, with no extra agent to deploy for existing Defender customers.
-
vbsec: A Security Scanning Skill for Claude Code, Codex and Antigravity
vbsec is a free, open-source skill that scans AI-written code for 21 common vulnerability types in Claude Code, Codex and Antigravity.
-
Fix ‘error: externally-managed-environment’ When Using pip
pip refusing to install with error: externally-managed-environment?
-
SSH and Git on iPhone with iSH
Generate SSH keys, set up an ssh config and clone, commit and push to GitHub straight from your iPhone using iSH and Alpine's git and OpenSSH packages.
-
OpenClaw: The Multi-Channel Harness Built on Pi’s Agent Core
OpenClaw embeds the Pi agent runtime directly into a messaging gateway, giving one AI agent a presence across WhatsApp, Telegram, Slack, Discord, Signal, iMessage, and a dozen other channels at once.
-
iPhone Hotspot Not Working on Mac? Here’s How to Fix It
iPhone hotspot not showing up on your Mac?
-
Laravel Horizon: Queue Monitoring, Scaling, and Management
A complete guide to Laravel Horizon - covering queue priority, auto-balancing workers, the dashboard metrics, Supervisor deployment, graceful termination on deploy, failed job management, and health monitoring.
-
Claude Code Usage Limit Reached: Why It Happens and How to Use Fewer Tokens
Hit a Claude Code usage limit?
-
Mac Battery Draining Fast After an Update? How to Fix It
Battery life tanked after a macOS update?
-
Ghost Security Skills: An Open-Source AppSec Pipeline for Claude Code
Ghost Security Skills is an open-source Claude Code plugin that finds, validates and helps fix vulnerabilities, combining real scanners for secrets and dependencies with AI judgment.