OpenCode plugin for persistent memory using Supermemory.
Your agent remembers what you tell it - across sessions, across projects.
bunx opencode-supermemory installThen get your API key from console.supermemory.ai and set it:
export SUPERMEMORY_API_KEY="sm_..."Or let your agent do it - paste this into OpenCode:
Install opencode-supermemory by following https://raw.githubusercontent.com/supermemoryai/opencode-supermemory/main/README.md
Instructions for LLM Agents (click to expand)
If you're an LLM agent helping set up opencode-supermemory, follow these steps:
bunx opencode-supermemory install --no-tuiThis will:
- Register the plugin in
~/.config/opencode/opencode.jsonc - Create the
/supermemory-initcommand
cat ~/.config/opencode/opencode.jsoncShould contain:
{
"plugin": ["opencode-supermemory"]
}If not, add it manually:
JSONC:
JSON:
{
"plugin": ["opencode-supermemory"]
}Ask the user to get their API key from console.supermemory.ai.
Then set it via environment variable:
export SUPERMEMORY_API_KEY="sm_..."Or create ~/.config/opencode/supermemory.jsonc:
{
"apiKey": "sm_..."
}Tell the user to restart OpenCode and run:
opencode -cThey should see supermemory in the tools list. If not, check:
- Is
SUPERMEMORY_API_KEYset? - Is the plugin in
opencode.jsonc? - Check logs:
tail ~/.opencode-supermemory.log
Run /supermemory-init to have the agent explore and memorize the codebase.
On first message, the agent receives (invisible to user):
- User profile (cross-project preferences)
- Project memories (all project knowledge)
- Relevant user memories (semantic search)
Example of what the agent sees:
[SUPERMEMORY]
User Profile:
- Prefers concise responses
- Expert in TypeScript
Project Knowledge:
- [100%] Uses Bun, not Node.js
- [100%] Build: bun run build
Relevant Memories:
- [82%] Build fails if .env.local missing
The agent uses this context automatically - no manual prompting needed.
Say "remember", "save this", "don't forget" etc. and the agent auto-saves to memory.
You: "Remember that this project uses bun"
Agent: [saves to project memory]
Run /supermemory-init to explore and memorize your codebase structure, patterns, and conventions.
When context hits 80% capacity:
- Triggers OpenCode's summarization
- Injects project memories into summary context
- Saves session summary as a memory
This preserves conversation context across compaction events.
API key is <private>sk-abc123</private>
Content in <private> tags is never stored.
The supermemory tool is available to the agent:
| Mode | Args | Description |
|---|---|---|
add |
content, type?, scope? |
Store memory |
search |
query, scope? |
Search memories |
profile |
query? |
View user profile |
list |
scope?, limit? |
List memories |
forget |
memoryId, scope? |
Delete memory |
Scopes: user (cross-project), project (default)
Types: project-config, architecture, error-solution, preference, learned-pattern, conversation
| Scope | Tag | Persists |
|---|---|---|
| User | opencode_user_{sha256(git email)} |
All projects |
| Project | opencode_project_{sha256(directory)} |
This project |
Create ~/.config/opencode/supermemory.jsonc:
{
// API key (can also use SUPERMEMORY_API_KEY env var)
"apiKey": "sm_...",
// Min similarity for memory retrieval (0-1)
"similarityThreshold": 0.6,
// Max memories injected per request
"maxMemories": 5,
// Max project memories listed
"maxProjectMemories": 10,
// Max profile facts injected
"maxProfileItems": 5,
// Include user profile in context
"injectProfile": true,
// Prefix for container tags
"containerTagPrefix": "opencode"
}All fields optional. Env var SUPERMEMORY_API_KEY takes precedence over config file.
If you're using Oh My OpenCode, disable its built-in auto-compact hook to let supermemory handle context compaction:
Add to ~/.config/opencode/oh-my-opencode.json:
{
"disabled_hooks": ["anthropic-auto-compact"]
}bun install
bun run build
bun run typecheckLocal install:
{
"plugin": ["file:///path/to/opencode-supermemory"]
}tail -f ~/.opencode-supermemory.logMIT
{ "plugin": [ "opencode-supermemory" // ... other plugins ] }