OpenCode - terminal AI with screen memory

Integrate screenpipe with OpenCode to give your terminal AI assistant access to your screen history, audio transcriptions, and app context.

OpenCode is a powerful terminal-based AI coding assistant. it implements the Agent Skills open standard, which means screenpipe skills work out of the box.

what is OpenCode?

OpenCode is a terminal-native AI coding assistant that brings Claude, GPT-4, and other AI models to your command line. with screenpipe integration, OpenCode gets context about what you’ve been working on across all your apps.

setup

OpenCode discovers skills from multiple locations. choose the option that works best for you:

project-level (current project only)

# copy to current project
mkdir -p .opencode/skills
cp -r /path/to/screenpipe/.claude/agents/* .opencode/skills/

OpenCode uses the same Agent Skills format as Claude Code. files in .claude/agents/ work in .opencode/skills/ and vice versa.

available skills

once installed, OpenCode has access to these screenpipe skills:

skill	description
screenpipe-query	search screen OCR, audio transcriptions, and UI events (keyboard input, clicks, app switches, clipboard)
screenpipe-health	check status, diagnose issues, verify permissions
screenpipe-logs	retrieve and analyze screenpipe logs

usage

OpenCode automatically discovers installed skills. invoke them in your prompts:

explicit skill invocation

# start OpenCode
opencode

> @screenpipe-query find what I was reading about docker compose

> @screenpipe-health check if recording is working

> @screenpipe-logs show me errors from today

automatic skill selection

let OpenCode choose the right skill automatically:

> what was I working on in VS Code this morning?
# OpenCode will invoke screenpipe-query

> is my screen recording working?
# OpenCode will invoke screenpipe-health

> show me recent screenpipe errors
# OpenCode will invoke screenpipe-logs

OpenCode’s AI decides which skill to use based on your query.

example workflows

context-aware coding:

> I was looking at a React component earlier that had a cool
> animation effect, use screenpipe to find it and help me
> implement something similar

OpenCode will:

1. invoke screenpipe-query to search your screen history
2. find the React component
3. analyze the code
4. help you implement it

recall documentation:

> use screenpipe to find that kubernetes docs page I was
> reading about pod scheduling, then explain it

meeting follow-up:

> what did we discuss in the team call about the database
> migration? use screenpipe to find it

debug from memory:

> there was an error in my terminal earlier, use screenpipe
> to find it and help me fix it

check screenpipe status:

> is screenpipe recording my screen properly?

MCP alternative

OpenCode also supports MCP servers if you prefer MCP over skills:

# add screenpipe via MCP
opencode config mcp add screenpipe "npx -y screenpipe-mcp"

skills vs MCP

	skills	MCP
token efficiency	✅ more efficient	❌ more overhead
setup	copy files	run command
recommended for	local tools, codebase context	external APIs, remote services
OpenCode team recommendation	✅ preferred	use for external integrations

the OpenCode team recommends skills for local tools like screenpipe. reserve MCP for external API integrations.

skill format reference

screenpipe skills follow the Agent Skills standard. here’s the structure:

---
name: screenpipe-query
description: Query screen recordings and audio transcriptions
tools:
  - Bash
  - WebFetch
---

# Screenpipe Query Agent

Instructions for querying screenpipe data...

you can customize these skills or create your own following the same format.

creating custom screenpipe skills

want to create your own screenpipe skill?

1. create a .md file in .opencode/skills/ or ~/.opencode/skills/
2. add YAML frontmatter with name, description, and tools
3. write instructions for the AI agent
4. save and test

example:

---
name: screenpipe-meetings
description: Analyze and summarize meeting transcriptions
tools:
  - Bash
---

# Screenpipe Meeting Analyzer

You help users analyze their meeting transcriptions from screenpipe.

## Available endpoints:

- Search audio: GET http://localhost:3030/search?content_type=audio

When the user asks about meetings:
1. Search audio transcriptions
2. Filter by time range
3. Summarize key points
4. Extract action items

requirements

• screenpipe running on localhost:3030
• OpenCode installed: go install github.com/opencode-ai/opencode@latest
• skills copied to .opencode/skills/ or ~/.opencode/skills/

troubleshooting

skills not discovered?

• run opencode skills list to see available skills
• verify files are in correct location: .opencode/skills/ or ~/.opencode/skills/
• check skill file ends in .md
• ensure YAML frontmatter is valid

queries returning no data?

• verify screenpipe is running: curl http://localhost:3030/health
• check data exists: curl "http://localhost:3030/search?limit=1"
• ensure screenpipe has screen recording permissions
• try a broader time range

OpenCode not using skills?

• mention the skill explicitly: @screenpipe-query find...
• check skill description matches your query intent
• ensure the skill file is properly formatted

skill invocation errors?

• check screenpipe API is accessible
• verify Node.js is installed (for npx commands in skills)
• look at OpenCode’s debug output for details

OpenCode installation issues?

• ensure Go is installed: go version
• try reinstalling: go install github.com/opencode-ai/opencode@latest
• check your $GOPATH/bin is in your $PATH

still stuck? ask in our discord — we can help debug setup issues.

resources

• OpenCode documentation
• Agent Skills specification
• screenpipe skills source
• MCP server alternative