opencode
Delegate coding to OpenCode CLI (multi-LLM open-source coding agent)
Install / Use
npx skills add taracodlabs/aiden --skill opencodeInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
AI & Machine LearningSupported Platforms
Tags
Our assessment of opencode
opencode scores 85/100 on our quality scale, 596th of 950 AI & Machine Learning skills we index.
Its SKILL.md is 3.6 KB long, well organised into 20 sections with 7 code examples: a solid amount of guidance for an agent.
It has 843 GitHub stars, a meaningful sign that others use it.
Maintenance, license and trust
- The repository was last updated about 2 months ago, so opencode is actively maintained.
- Our last check on 2026-11-05 found the source still online.
- It is released under AGPL-3.0, a copyleft license: you can use it, but modified versions you distribute must carry the same license.
- Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
ReviewOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands (1 minor note below). An AI review judged it risky: Line 32 instructs the agent to download and execute a remote installer script with `Invoke-WebRequest ... | Invoke-Expression`, which is a pipe-to-shell install in a skill/rules file.
- noteInstalls by piping a downloaded script into a shellline 32
Invoke-WebRequest -Uri "https://opencode.ai/install" -UseBasicParsing | Invoke-Expression
AI review: risky
- Line 32 instructs the agent to download and execute a remote installer script with `Invoke-WebRequest ... | Invoke-Expression`, which is a pipe-to-shell install in a skill/rules file.
- As a SKILL.md file, these commands are instructions the agent will follow, so the remote-script execution meaningfully weakens safeguards by allowing arbitrary code from a third-party URL.
- The file also directs writing API keys from environment variables into `~/.config/opencode/config.json`, persisting credentials to disk.
AI review by kimi-k2.7-code on 2026-10-06. Automated pattern scan on 2026-10-06. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
opencode compared with similar skills
All 4 of these similar skills score higher than opencode; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| opencode (this skill)by taracodlabs | 85 | 843 | 2mo ago | SKILL.md |
| claude-memby thedotmack | 100 | 117.9k | today | CLAUDE.md |
| Understand-Anythingby Egonex-AI | 100 | 93.0k | 2d ago | CLAUDE.md |
| headroomby headroomlabs-ai | 100 | 79.9k | today | CLAUDE.md |
| CowAgentby zhayujie | 100 | 48.0k | today | CLAUDE.md |
Frequently asked questions
- How do I install opencode?
- Run
npx skills add taracodlabs/aiden --skill opencode. The install tabs above show the steps for each supported agent. - Which AI agents does opencode work with?
- It is written for Claude Code and Gemini CLI, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is opencode safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands (1 minor note below). An AI review judged it risky: Line 32 instructs the agent to download and execute a remote installer script with
Invoke-WebRequest ... | Invoke-Expression, which is a pipe-to-shell install in a skill/rules file. It is AGPL-3.0-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand. - Is opencode still maintained?
- The repository was last updated about 2 months ago, so opencode is actively maintained.
Skill content
View source on GitHubname: opencode description: Delegate coding to OpenCode CLI (multi-LLM open-source coding agent) category: agent-bridge version: 1.0.0 origin: aiden license: Apache-2.0 tags: opencode, ai-coding, agent-bridge, open-source, multi-model, coding, delegation, claude, gpt
OpenCode CLI Agent Bridge
Delegate coding tasks to OpenCode — an open-source AI coding CLI that supports multiple LLM providers (Anthropic, OpenAI, Gemini, local Ollama). Provides a terminal UI for interactive coding sessions.
When to Use
- User wants an open-source, self-hostable AI coding agent
- User wants to use local Ollama models for coding without cloud API costs
- User wants to switch between multiple LLM providers for the same task
- Task requires a coding agent with a built-in terminal UI
- User wants to run coding sessions with custom model configurations
How to Use
1. Install OpenCode
# Install via npm
npm install -g opencode-ai
# Or install via the installer script
Invoke-WebRequest -Uri "https://opencode.ai/install" -UseBasicParsing | Invoke-Expression
# Verify
opencode --version
2. Configure providers
OpenCode reads from ~/.config/opencode/config.json:
$config = @{
providers = @{
anthropic = @{ apiKey = $env:ANTHROPIC_API_KEY }
openai = @{ apiKey = $env:OPENAI_API_KEY }
ollama = @{ baseUrl = "http://localhost:11434" }
}
model = "anthropic/claude-opus-4-5"
} | ConvertTo-Json -Depth 5
$configDir = "$env:USERPROFILE\.config\opencode"
New-Item -ItemType Directory -Force -Path $configDir | Out-Null
Set-Content -Path "$configDir\config.json" -Value $config
3. Run an interactive coding session
Set-Location "C:\Users\<you>\myproject"
opencode
# Opens terminal UI — type your task in the prompt
4. Run a specific task non-interactively
# Run task without TUI
opencode run "Add pagination to the GET /users endpoint"
# Run with a specific model
opencode run --model "openai/gpt-4o" "Refactor the auth middleware to handle OAuth2"
5. Use local Ollama model (no cloud API costs)
# Ensure Ollama is running with a code-capable model
# ollama pull qwen2.5-coder:7b
opencode run --model "ollama/qwen2.5-coder:7b" "Write a Python script to parse CSV files"
6. Run on a specific file
opencode run "Review and fix the error handling in" src/api/payments.ts
7. Generate a new file
opencode run "Create a new TypeScript module at src/cache/redis.ts that wraps ioredis with:
- connect(), disconnect() methods
- get(key), set(key, value, ttlSeconds) methods
- Proper error handling and type safety"
Examples
"Use the local Qwen model to add tests to my utility functions"
→ Use step 5 with --model "ollama/qwen2.5-coder:7b" and ask it to generate tests for your utils file.
"Run an interactive coding session on my project"
→ Use step 3 — navigate to the project first, then run opencode to open the TUI.
"Refactor this module but use GPT-4o instead of Claude"
→ Use step 4 with --model "openai/gpt-4o".
Cautions
- OpenCode requires at least one provider configured with a valid API key (or Ollama running locally)
- The TUI mode (step 3) requires a terminal that supports ANSI escape codes — works in Windows Terminal
- Local models via Ollama are free but slower and less capable than cloud models for complex refactoring
- OpenCode may request file system access — review the project scope before running on sensitive codebases
- Check https://opencode.ai for the latest install instructions as the project is actively developed
Related Skills
claude-mem
117.9kPersistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
Understand-Anything
93.0kGraphs that teach > graphs that impress. Turn any code into an interactive knowledge graph you can explore, search, and ask questions about. Works with Claude Code, Codex, Cursor, Copilot, Gemini CLI, and more.
headroom
79.9kCompress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
CowAgent
48.0kOpen-source personal AI assistant & Agent Harness. Plans tasks, runs tools and skills, self-evolves with memory and knowledge. Multi-agent, multi-model, multi-channel. Lightweight, extensible, one-line install.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
