


Founder at @qbashdev- an AI Operations Platform. Founding Partner at Isle of Code - est. 2012.


How is this different than dependabot? Dependabot tells you a vulnerable package is in your lockfile, but that doesn't mean your code actually calls it. That is gated behind an Enterprise plan. This helps cut out noise (of which there's a lot rn).
CVEs published this year have already passed 100k (VulnCheck). This is more than 2x 2025. Here's how to build a qbash Task that sorts through the noise and opens a Slack thread per advisory that impacts you, and takes it to a merged fix or a written decision. This works on any level of Github plan.

Every time your coding agent pushes to GitHub, it can trigger your GitHub Actions, and costs climb fast. qbash Tasks let you program functions that connect to GitHub to triage and cut your Actions minutes. This one found 50+ superseded runs in a week. qbash.com/github-actions-cos…

my spidey senses are telling me the labs are, as we speak, using unguardrailed versions of their unreleased models to systematically reverse-engineer ALL of the internet’s software, apps, and games, then feeding the reconstructed source code straight into their training pipelines while simultaneously building guardrails to ladder-pull the rest of us


@tessi3r had Claude Code build a qbash Task over the official qbash MCP that posts a video to YouTube, LinkedIn and X with copy for each platform. Now his agent calls that Task as a tool. Add mcp.qbash.com/mcp

Federal Reserve data in Claude via the qbash MCP. Add your FRED API key, then ask in plain English. Claude pulls the series, charts inflation since 2020 against the Fed Funds rate and checks the release calendar. mcp.qbash.com/mcp

the official qbash MCP in Claude: 1. add mcp.qbash.com/mcp 2. connect Drive, ask what's low in inventory 3. turn the 4-week reorder into a Python Task 4. connect Slack, post it to a channel 5. schedule for 9am weekdays runs with Claude closed. qbash.com/docs/mcp-servers/o…

Here's how to connect Muse to qbash's MCP Server - Unlocking Durable Task Building, Memory Sharing and a plethora of Connector capabilities that can be distributed across teams across all major LLM providers including Muse, Claude, ChatGPT, Grokbot and more.

On Monday we took the qbash MCP server live - build multi model tasks, call connectors, write functions to keep agents on the rails: qbash.com/blog/release-notes…

Quick tutorial connecting qbash MCP servers to Grokbot. Use these to share memories across users/sessions, distribute otherwise unavailable connectors or share deterministic AI tasks with a team spread across Grokbot, Muse, Codex, Claude, etc: youtu.be/i4emsHOFraQ

Nobody owns which agents keep running or who answers when they break. Some call that role AI ops. Most of the job is moving agent work into fixed tasks. qbash.com/blog/who-owns-the-…

This week on qbash : Muse + Grokbot Support, a heavily improved Task Builder AI and build custom MCP servers with any Connector/Task/Memory qbash.com/blog/release-notes…

As requested, I built a Jev + Claude loop in @qbashdev to conquer AI slop. Long story short: Don't try it.

Seeing people solve problems with Grokbot + Muse this week has been fun. But what happens when they need to repeat the solve? I still lean towards agentic being used to discover the answer to a problem but if there is a pattern I move deterministic pretty quickly: qbash.com/blog/grok-bot-muse…
