» Tag
cli
54 postsDeja: Open-Source Local Memory Layer for Coding Agents
Deja indexes Claude Code, Codex CLI and opencode session logs into a local, redacted memory layer with MCP recall, auto-context, and SSH sync.
Telegram Serverless: Run Bot Backends Without Servers
Telegram Serverless runs bot and Mini App backend code in an isolated V8 sandbox with no infrastructure, using a CLI for deploys and database migrations.
MCP vs CLI: What Actually Survived in Production
A retrospective on MCP vs CLI for AI agents: token costs, lifecycle independence, containers, and a decision tree built from production failures.
CommitBrief — AI code reviews, right in your terminal
A provider-agnostic, local-first CLI that reviews your staged changes, a historic range, or a whole GitHub pull request. Zero telemetry, no server. Free and open source.
commitbrief.comTermDOM: HTML, CSS, and JavaScript for Terminal Apps
TermDOM enables the creation of interactive interfaces using HTML, CSS, and JavaScript for terminal applications.
Progress Bar Is Not an API: Separating CLI UI from Events
Co-op Translator v0.20.0 separates its Rich CLI UI from a structured event stream, freeing integrations from parsing fragile console text.
Kiro CLI Performance Degrades After an Hour: Here's How to Fix It
Discover why long sessions in Kiro CLI cause problems and how to address them.
Analysis of What xAI Grok Build CLI Sends
A detailed analysis of what xAI Grok Build CLI sends, including important findings about secret information and file contents.
aeovim: Orchestrating Claude Code agents like Neovim
aeovim is a Rust TUI that applies Neovim's modal, split-based workflow to LLM coding agents, letting developers orchestrate multiple Claude Code sessions from the keyboard.
Mitchell Hashimoto on Ghostty, Zig, and Terminal Protocols
HashiCorp founder Mitchell Hashimoto explains why he built Ghostty in Zig, the limits of the legacy PTY protocol, and his ideas for new terminal screen and button protocols.
DevTime scans Cal.com and admits uncertainty instead of guessing
Local-first scanner DevTime flagged Cal.com's billing webhooks as unproven behavior rather than guessing from file names, exposing why evidence-based confidence matters for AI coding agents.