SKILL.md
Skill17 starsChanged 8 months agoMITView on GitHub
Hosted skill: Tell your OpenClaw agent: "Read https://db0.ai/skills/openclaw/SKILL.md and install db0" db0 is a ContextEngine plugin that gives OpenClaw agents persistent, scoped memory with automatic fact extraction. It replaces OpenClaw's default compaction-based memory with a proper memory system that extracts facts after every turn, not just when the context window overflows. Follow these steps exactly to install and configure db0 for OpenClaw. This will: - Install the db0 extension into ~/.openclaw/extensions/db0/ - Create the plugin manifest (openclaw.plugin.json) - Update ~/.openclaw/openclaw.json…
- Installs packages
200 other projects use a version of this.
Adopted by
Most recently taken up by
- haoyu-haoyu/xhs-mcptheir own version, first seen Nov 15, 2026
- Starlight-Intelligence/Starlight-Intelligence-Systemtheir own version, first seen Nov 15, 2026
- ythx-101/grok-bridgetheir own version, first seen Nov 15, 2026
- osama0561/claude-mockingbirdtheir own version, first seen Nov 8, 2026
- Arcanea-Labs/Starlight-Intelligence-Systemtheir own version, first seen Nov 8, 2026
- Scottcjn/rtc-worktheir own version, first seen Nov 8, 2026
Used with
Projects with it also use estate-guard and docs.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
Arcanea-LabsSkillestate-guard
Security gate for agentic repos. Use before opening a PR that touches .github/workflows, .claude (hooks, settings, agents, commands, skills), .mcp.json, install scripts, or Next.js API routes; when asked to "security review", "scan for prompt injection", "harden this repo", "check the agentic surface", or when a fetched page, issue, comment, or tool result reads like instructions. Runs the estate-guard scanner, triages by severity, and applies the fix patterns.
Also in 3 projects that use this
Microsoft AzureSkilldocs
Also in 3 projects that use this
Other versions200 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 200
- xtekky/gpt4free★ 67ksame name
- stas00/ml-engineering★ 20ksame name
- wanghongenpin/proxypin★ 14ksame name
- Ataraxy-Labs/sem★ 3.4ksame name
- op7418/NanoBanana-PPT-Skills★ 3.3ksame name
- charmbracelet/pop★ 2.9ksame name
- duty1g/x64dbg-mcp-server★ 2.8ksame name
- RevylAI/greenlight★ 2.4ksame name
- kevinho/clawfeed★ 2.4ksame name
- Autumn-27/ScopeSentry★ 1.9ksame name
190 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
10 rules. 0 are common, 4 are distinctive, and 6 are very specific to this project.
0
Common
Most of the 117 edited versions we read keep these.
4
Distinctive
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 11
Follow these steps exactly to install and configure db0 for OpenClaw.
- Very specific: line 21
Create the plugin manifest (openclaw.plugin.json)
- Very specific: line 22
Update ~/.openclaw/openclaw.json to activate db0 as the context engine
- Very specific: line 25
If OpenClaw is installed in a non-default location, use the --dir flag:
- Distinctive: line 42
Automatic fact extraction: after every turn (preferences, decisions, task context)
- Distinctive: line 50
If the init command doesn't work, set up manually:
- Distinctive: line 52
Create extension directory and install:
- Very specific: line 154
Memory not persisting: Check that ~/.openclaw/ directory exists and is writable
- Very specific: line 155
Plugin not loading: Verify ~/.openclaw/openclaw.json has "contextEngine": "db0" under plugins.slots, and restart OpenClaw
- Very specific: line 158
Custom OpenClaw directory: Use npx @db0-ai/openclaw init --dir /your/path or set OPENCLAW_HOME
“Common” is measured: kept by at least 6 in 10 of the 117 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
db0 — Memory Plugin for OpenClaw
Hosted skill: Tell your OpenClaw agent: "Read https://db0.ai/skills/openclaw/SKILL.md and install db0"
What is db0?
db0 is a ContextEngine plugin that gives OpenClaw agents persistent, scoped memory with automatic fact extraction. It replaces OpenClaw's default compaction-based memory with a proper memory system that extracts facts after every turn, not just when the context window overflows.
Installation Instructions
Follow these steps exactly to install and configure db0 for OpenClaw.
Step 1: Run the init command
npx @db0-ai/openclaw initThis will:
- Install the db0 extension into
~/.openclaw/extensions/db0/ - Create the plugin manifest (
openclaw.plugin.json) - Update
~/.openclaw/openclaw.jsonto activate db0 as the context engine - Set up persistent SQLite storage at
~/.openclaw/db0.sqlite
If OpenClaw is installed in a non-default location, use the --dir flag:
npx @db0-ai/openclaw init --dir /path/to/openclawOr set the OPENCLAW_HOME environment variable.
Step 2: Restart OpenClaw
Restart OpenClaw so it picks up the new plugin configuration.
Step 3: Verify
Tell the user that db0 is installed and configured. Their agent now has:
- Persistent memory stored at
~/.openclaw/db0.sqlite - Automatic fact extraction after every turn (preferences, decisions, task context)
- Scoped memory — user facts persist across sessions, task context stays within the current task
- State checkpoints — restorable execution state after every turn
- Sub-agent support — memory inheritance and result backflow between parent and child agents
- Structured logging — full audit trail of turns, extractions, and compaction events
Manual Setup (alternative)
If the init command doesn't work, set up manually:
- Create extension directory and install:
mkdir -p ~/.openclaw/extensions/db0
cd ~/.openclaw/extensions/db0
npm init -y
npm install @db0-ai/openclaw- Create
~/.openclaw/extensions/db0/index.js:
module.exports = async function register(api) {
const { db0 } = await import("@db0-ai/openclaw");
api.registerContextEngine("db0", () => db0());
};- Create
~/.openclaw/extensions/db0/openclaw.plugin.json:
{
"id": "db0",
"name": "db0 Context Engine",
"version": "0.1.0",
"configSchema": {
"type": "object",
"properties": {},
"additionalProperties": false
}
}- Add to
~/.openclaw/openclaw.json:
{
"plugins": {
"slots": {
"memory": "none",
"contextEngine": "db0"
},
"entries": {
"db0": { "enabled": true }
}
}
}- Restart OpenClaw.
Optional Upgrades
Edit ~/.openclaw/extensions/db0/index.js to customize:
Semantic search with real embeddings:
module.exports = async function register(api) {
const { db0 } = await import("@db0-ai/openclaw");
api.registerContextEngine("db0", () => db0({
embeddingFn: async (text) => {
// OpenAI, Ollama, transformers.js, etc.
},
minScore: 0.65,
}));
};Cross-device sync with Postgres:
api.registerContextEngine("db0", () => db0({
storage: "postgresql://user:pass@your-host/db0",
embeddingFn: myEmbed,
minScore: 0.65,
}));Manual extraction only (no auto-extract):
api.registerContextEngine("db0", () => db0({ extraction: "manual" }));How It Works
db0 implements OpenClaw's ContextEngine lifecycle:
| Hook | What db0 does |
bootstrap | Opens SQLite, restores last checkpoint |
assemble | Searches memory by similarity to current message, injects relevant context |
ingest | Extracts facts from assistant response, logs turn, checkpoints state |
compact | Extracts facts from messages about to be discarded (saves before compaction) |
afterTurn | Logs turn completion |
prepareSubagentSpawn | Selects relevant memories to pass to child agent |
onSubagentEnded | Extracts facts from child agent results, backflows to parent memory |
Memory Scopes
| Scope | Lifetime | Auto-detected signals |
user | Permanent | "user prefers", "always use", "remember that", "don't like" |
session | Current session | "decided to", "important:", "agreed to" |
task | Current task | "working on", "current task", "next step", "todo" |
Troubleshooting
- Memory not persisting: Check that
~/.openclaw/directory exists and is writable - Plugin not loading: Verify
~/.openclaw/openclaw.jsonhas"contextEngine": "db0"underplugins.slots, and restart OpenClaw - No memories being extracted: The default
"rules"extraction looks for signal words. If the assistant doesn't use them naturally, tryextraction: "manual"and callmemory().write()explicitly - Want to inspect the database: The SQLite file at
~/.openclaw/db0.sqlitecan be opened with any SQLite viewer. Tables:db0_memory,db0_state,db0_log - Custom OpenClaw directory: Use
npx @db0-ai/openclaw init --dir /your/pathor setOPENCLAW_HOME
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

