Tool
OpenCode
Agentic coding CLI that runs in your terminal.
Affiliate link. We may earn a commission if you sign up — at no cost to you. See our disclosure for details.
About OpenCode
OpenCode is a terminal-based AI coding agent that edits files, runs commands, and iterates on tasks inside your existing projects. Unlike cloud IDEs, it works locally with your own shell, git history, and LSP, so you keep your editor and workflow. We tested it against multi-file refactors, test generation, and dependency upgrades — it writes, runs tests, and rolls back on failure without leaving the repository. If you already use Claude Code or similar terminal agents, OpenCode is a strong alternative with a generous free tier and transparent per-token billing.
Articles featuring OpenCode
-
OpenAI Deployment Layer: The Assistants API Precedent
OpenAI shipped the Assistants API in November 2023 and marked it for sunset 16 months later. That precedent is how to price the new deployment stack.
2026-09-02 · infrastructure
-
DeepSeek MLA: 70 GB of KV Cache at 1M Tokens
No DeepSeek-V4 config is public yet. The V3 one is, and its KV-cache math tells you what a million-token window actually costs in GPU memory.
2026-09-02 · ai-dev-tools
-
Your Bot Filter Misses Crawlers That Send a Referer
A no-referer heuristic passed 234 of 506 affiliate clicks as human. A country exclusion cut the same set to 117.
2026-08-21 · dev-knowledge
-
A Jaccard Gate at 0.50 Before the Model Runs
Why we strip year tokens before scoring, what the length>2 rule silently breaks, and when to switch to embeddings.
2026-08-21 · dev-knowledge
-
Bing Webmaster API's 100-URL Cap: 289 URLs, Three Days
Quota is 100 a day against 1300 a month, GetQueryStats was still empty at the end, and two error shapes will kill a cron job.
2026-08-21 · infrastructure
-
Astro middleware can't serve 410 under output: 'static'
We deleted 434 articles. Middleware runs at build time and 404s in production; two prerender:false routes fix it on Cloudflare Workers.
2026-08-19 · infrastructure
-
Workers KV's 60-Second Consistency Window
It's a cache TTL you cannot lower, not a replication delay. It bites cached nulls, uniqueness checks, and multi-key updates. When to swap in a Durable Object.
2026-08-17 · infrastructure
-
Automations Worth Building First
A sort key for your automation backlog from running scheduled agents in production: script it, agent it, or leave it a checklist -- plus three failure modes.
2026-08-13 · ai-knowledge-work
-
A Daily LLM Digest Agent Costs $166 and 34 Hours a Year
A year of one scheduled digest agent in production: token costs per stage, infrastructure line items, and the 41 runs that needed a human.
2026-08-13 · ai-dev-tools
-
Cache Races Make IndexNow Miss Pages You Just Shipped
A deploy API returning 200 does not mean the new URLs are reachable at the edge. Here is the verification sequence that stops wasted submissions.
2026-08-13 · infrastructure
-
When Agent Output Passes Zod and Still Breaks the Build
Four failure classes that survive a clean parse, and the three-layer validation pass we run instead.
2026-08-13 · ai-dev-tools
-
Idempotency: Publishing Agents That Survive a Mid-Run Kill
A per-channel ledger makes cross-posting resumable. Why exit code 0 is not a receipt, and what to do when an API has no idempotency key.
2026-08-13 · ai-dev-tools
-
Your Automation's Numbers Are Probably Wrong: 89% Bot Clicks
Two production metric failures from running scheduled agents: an upsert that overwrote instead of incrementing, how we found both, and the checks we run now.
2026-08-13 · meta
-
Cron Agents Die Silently: 4 Failure Modes
They exit 0 and do nothing. What that looks like when running scheduled agents in production, and the three assertions that catch them.
2026-08-13 · meta
-
AI Agent Auth: API Keys vs Device Flow vs Scoped Tokens
Three credential models for non-human callers: static keys, the OAuth 2.0 device grant, and short-lived scoped tokens - and when each one fits.
2026-08-13 · ai-dev-tools
-
Error Messages as an Agent Interface
A field-by-field guide to API error bodies: stable codes, retryable flags, wait hints, fix examples, and the shapes that trap agents in retry loops.
2026-08-13 · ai-dev-tools
-
Agent Experience (AX): When Your User Is an AI Agent
What breaks when AI agents use your product: auth, docs, error messages, and state handling - and the order to fix them.
2026-08-13 · dev-knowledge
-
Auditing a 660-Article Archive for Near-Duplicate Content
The three-pass process: hashing, MinHash shingles, then embedding similarity, plus the rules for merging, differentiating, or keeping overlapping posts.
2026-08-12 · meta
-
Object Storage Lifecycle Policies: Cut Cost, Keep Data
Audit an S3-compatible bucket, then write lifecycle rules that avoid transition fees, minimum-duration charges, and versioning traps that raise bills.
2026-08-12 · infrastructure
-
Writing a Spec an AI Agent Can Actually Execute
Ground truth files, an interface contract, one acceptance command, and explicit out-of-bounds rules -- so the agent runs end to end without babysitting.
2026-08-12 · ai-dev-tools
-
MCP Server Security: What a Local Server Can Actually Reach
It runs as a child process with your user's permissions. Four commands to see which files it opens and hosts it dials, plus three containment changes.
2026-08-12 · ai-dev-tools
-
How Agent Tool-Use Loops Work
AI coding agents follow a loop of planning, reading, editing, and verifying. Understanding that loop helps you write prompts that get better results.
2026-07-16 · dev-knowledge
-
AI Agents for Code Review and Knowledge Sharing
AI coding agents can act as a first-pass code reviewer and knowledge capture tool. Here is how teams are using them without replacing human judgment.
2026-07-16 · ai-knowledge-work
-
AI Coding Agents for Data Analysis Workflows
OpenCode and similar agents can write scripts, clean data, and build charts. Where they help analysts, and where human judgment is still required.
2026-07-16 · ai-knowledge-work
-
Context Windows in AI Coding Agents Explained
A context window caps how much code an agent sees at once, which is why agents miss files, repeat work, and lose track of the task.
2026-07-16 · dev-knowledge
-
How AI Coding Agents Read and Index Your Codebase
AI agents do not magically understand your project. They use a mix of file listing, search, and context loading to find relevant code. Here is how it works.
2026-07-16 · dev-knowledge
-
Inline Completion vs Agentic Coding: What Is the Difference?
Inline completion and agentic coding are both AI-assisted development, but they solve different problems. Here is how to tell which one you need.
2026-07-16 · dev-knowledge
-
How to Measure Cost Per Task with Terminal AI Agents
OpenCode and Claude Code bill by the token, so cost per task is the metric that matters. Here is how we tracked it and what we learned.
2026-07-16 · ai-dev-tools
-
Multi-Agent Terminal Workflows with OpenCode and Aider
How we split work across OpenCode, Claude Code, and Aider in one terminal without losing track of which agent changed what.
2026-07-16 · ai-dev-tools
-
OpenCode for Documentation and Technical Writing Teams
Technical writers and documentation teams can use OpenCode to extract API changes, update code examples, and keep docs in sync with the codebase.
2026-07-16 · ai-knowledge-work
-
OpenCode First Project Setup: Install to First Passing Test
Configure providers, context files, and project conventions so the agent produces useful output on an existing codebase from day one.
2026-07-16 · ai-dev-tools
-
OpenCode for Technical PMs: Review Code Without Writing It
OpenCode's terminal agent explains diffs, traces changes, and answers codebase questions without replacing your engineers' workflow.
2026-07-16 · ai-knowledge-work
-
Running OpenCode with Local LLMs for Private AI Coding
We ran OpenCode against a local Ollama model on a proprietary codebase, with no source code sent to a cloud API.
2026-07-16 · ai-dev-tools
-
OpenCode Review: A Terminal-Native AI Coding Agent
We ran it as a daily driver for two weeks. It edits files, runs tests, and supports multiple models without replacing your IDE.
2026-07-16 · ai-dev-tools
-
Terminal AI Agents for Non-Coding Technical Tasks
AI coding agents are not just for writing code. They can write scripts, query logs, and automate terminal tasks that technical workers deal with every day.
2026-07-16 · ai-knowledge-work
-
Why Terminal-Based AI Agents Use Text In, Text Out
The terminal seems old-fashioned, but its text-based interface is exactly why AI coding agents work there. Here is the engineering reason behind the trend.
2026-07-16 · dev-knowledge
-
Aider Review: Open-Source AI Pair Programming for Any LLM
I tested Aider on 9 projects with 6 LLMs over six weeks for $47.30. Why git-native beats accept/reject buttons, and where terminal-only falls short.
2026-05-28 · ai-dev-tools
-
Macchiato Day 2: Live Token Metrics and Parallel Terminals for Claude Code and OpenCode
Macchiato's Day 2 update lands a live token/cost sidebar, consumption dashboards, and keyboard shortcuts for jumping between Claude Code and OpenCode in one terminal. Here is what shipped and who should care.
2026-05-26 · saas-productivity
-
Macchiato Day 2: Live Token Metrics and Parallel AI Terminals Reviewed
Macchiato's day-2 build adds a live token/cost sidebar and keyboard shortcuts for swapping between Claude Code and OpenCode in one terminal. Here's what shipped and what it means.
2026-05-26 · ai-dev-tools
-
GitHub Copilot Desktop vs Claude Code vs Codex CLI
GitHub's standalone Copilot desktop app changes the matchup. We compare workflow surface, approval semantics, and model neutrality.
2026-05-18 · ai-dev-tools
-
Cursor vs GitHub Copilot: Which Ships Faster in 2026?
We tested both on a Next.js app, a Python CLI, and a Rust library migration. Cursor won on velocity -- but one scenario still favors Copilot.
2026-05-14 · ai-dev-tools