OmniPlan and MCP: Planning a Project in Plain Words Through a Coding Agent
I plan projects on Gantt charts in OmniPlan. Through MCP a coding agent creates the tasks there, and I just tell it what I need.
I plan projects on Gantt charts in OmniPlan. Through MCP a coding agent creates the tasks there, and I just tell it what I need.
I built gcalc-mcp, a Model Context Protocol server in Go that wraps the battle-tested gcalcli and exposes Google Calendar to any AI assistant over stdio, with a hardcoded allowlist, confirm-gated deletes, and zero stored credentials.
I built imagemagick-mcp, a tiny, fast Model Context Protocol server in Go that wraps ImageMagick so any MCP-aware agent can crop, resize, and blur-fill images to exact dimensions, without ever distorting them.
I built jira-mcp, a Model Context Protocol server in Go that gives Claude, Cursor, and any MCP client a real handle on Jira: search with JQL, create and edit work items, run sprints, all from chat, riding on the Atlassian CLI you already log into.
I built kie-ai-mcp, a Model Context Protocol server in Go that bridges to the Kie AI image generation API: it creates tasks, polls them to completion, downloads the result to disk, and runs batches in parallel.
I built slack-mcp and slack-cli, a Model Context Protocol server and the Go CLI engine behind it that let Claude, Cursor, and any MCP client search Slack, unroll threads, and resolve who said what, with credentials encrypted at rest and no Slack app to approve.
I am available for contract-based collaboration. If you have an interesting project idea, schedule a call via Calendly.
Schedule a 30-min call