Notes from the coding terminal.
Product launches, feature notes, and engineering decisions behind a local-first AI coding runtime.
Every note starts from a surface you can inspect: a terminal view, a release, a workflow, or a boundary.
9 posts · newest first
All posts

Image generation belongs inside the coding session
An implementation preview of MendCode Image Gen: gpt-image-2 generation and editing through subscription OAuth, bounded references, durable PNG artifacts, and no separate API-key workflow.
Read articleWorkflows need their own runtime
A development preview of MendCode dynamic workflows: finite execution graphs, hardened loop scheduling, and graph memory that remain connected without becoming the same system.
workflows · loops · graph · memory
MendCode is the anti-chatbot harness for the loop era
Welcome to MendCode: a local-first AI coding harness built around agents, loops, permissions, memory, packages, review, Usage Insights, and a terminal surface developers can actually own.
launch · agents · loops · sovereignty
The prompt is part of the runtime
Minimal, focus, and full are runtime contracts. MendCode combines prompt context, repo instructions, conditional tool calls, and packages without turning every request into a mega prompt.
context · prompts · tools · packages
Own your cockpit: the coding terminal is product surface now
Themes are not enough. Agentic work needs profiles, widgets, custom pages, package-shaped workflows, and a terminal surface that matches how the team actually thinks.
customization · tui · widgets · profiles
Memory without lore: agents need reviewable continuity
Agent memory is useful only when the user can inspect, scope, approve, reject, and clean it up. Otherwise it becomes invisible lore that bends every future session.
memory · dream · review · continuity
Founder economics need local insight, not creepy analytics
AI agents make small teams capable of directing far more work. Usage Insights should help them operate that machine locally without turning developer behavior into surveillance.
usage · privacy · founder-economics · local-first
Packages are how teams stop copy-pasting their operating system
The good prompt is not enough. MendCode packages make workflows composable: skills, profiles, prompts, widgets, pages, context, themes, MCP, and policy in one inspectable unit.
packages · skills · workflows · ecosystem
Permission theater sucks. Agent speed needs real boundaries.
As agents get faster, permission prompts cannot be vague interruptions. They need blast-radius context, smart review, and visible boundaries inside the coding surface.
permissions · smart-approve · security · agents







