CLI AI AgentsManaging Reusable Prompts for Terminal WorkflowsRetyping your best prompt from memory loses its refinements every time. Managing cli agent reusable prompts as named, parameterized, versioned assets keeps the prompt quality you earned — and lets you share it.September 19, 2026·9 min read
CLI AI AgentsDistributing System Prompts With Your CLI ToolHardcoding your agent's system prompt as a string is the wrong place for it. Treating cli agent system prompt distribution as content — versioned, overridable, updatable — is how prompts evolve independently of code.September 19, 2026·9 min read
CLI AI AgentsBuilding a Plugin System for Your CLI AgentHow do you let people add tools to your agent without forking it? A cli agent plugin system lets users extend the agent with their own tools. Here's how to rebuild a hardcoded tool list into a real plugin system.September 19, 2026·8 min read
CLI AI AgentsHandling Long-Running Tasks in a CLI AgentA twenty-minute agent job you can't watch, interrupt, or resume is a gamble. Handling cli agent long running tasks well means checkpointing, safe interruption, and resumability — here's how a team did it.September 19, 2026·9 min read
CLI AI AgentsColorized, Readable Output for CLI AgentsThe change that most improves how a terminal agent feels isn't the model — it's color. But naive cli agent output formatting breaks in pipes and CI. Here's how to colorize output the right way.September 19, 2026·9 min read
CLI AI AgentsBuilding a CLI Agent That Runs Tests AutomaticallyAn agent that edits code but doesn't run tests is guessing. Making a cli agent auto run tests after each change turns claims into verified results — here's the edit-verify-iterate loop.September 17, 2026·9 min read
CLI AI AgentsLogging CLI Agent Actions for AuditLogging what the model said is not an audit trail. Real cli agent action logging records what your code executed — here's how to rebuild narration logging into structured, queryable audit records.September 17, 2026·9 min read
CLI AI AgentsUndo and Rollback in a CLI AgentWhat happens when your agent makes a change you didn't want? Good cli agent undo changes support turns a scramble into one command — here's why in-memory undo fails and git-backed checkpoints work.September 17, 2026·8 min read
CLI AI AgentsBuilding a Read-Only Mode for Safe ExplorationA cli agent read only mode is a hard, code-enforced guarantee that an agent can look but not touch. Here's how to build it as a real boundary, default it safe, and cover shell and network side effects.September 17, 2026·9 min read
CLI AI AgentsRate Limiting and Retries in a CLI AgentBy the time a rate-limit error reaches your code, the SDK already retried it. Good cli agent retry logic tells a 429 from a 529, honors retry-after, and avoids the double-retry deadlock.September 17, 2026·9 min read
CLI AI AgentsPackaging a CLI Agent for DistributionA broken first npm publish leaks keys, ships 40MB, and installs no command. Here's how to package a CLI agent npm users can install and trust — plus secure publishing with provenance in 2026.September 17, 2026·9 min read
CLI AI AgentsTesting a CLI AI Agent"You can't test AI, it's non-deterministic" is wrong. When you test a CLI AI agent you're testing the harness, not the model — here's the two-layer strategy that stops shipping broken agents.September 17, 2026·9 min read
