skill
discord-soul
Turn your Discord server into a living agent with memory, personality, and identity
Set up
Skills are installed by placing their folder where your agent loads skills.
The download comes from github.com/kgeesawor/discord-soul itself, not from our server, so what you get is exactly what its author published.
Install instructions come from the README
README
synced Oct 26, 2026Discord Soul
Turn your Discord server into a living, breathing agent that embodies your community's identity, memory, and culture.
Built for OpenClaw — the open-source AI agent framework.
What You Get
An agent that:
- Remembers every conversation
- Speaks in your community's voice
- Knows the key figures and inside jokes
- Grows as new messages arrive daily
- Answers questions about community history
Quick Start
# 1. Create agent from your Discord
./scripts/create_agent.sh --name "my-community" --guild YOUR_GUILD_ID
# 2. Add to OpenClaw config and restart
# 3. Set up daily updates (cron)
./scripts/update_agent.sh --agent ./my-agent --db ./discord.sqlite --guild YOUR_GUILD_ID
The Process
- Export — Pull all messages from Discord
- Security — Filter prompt injections (regex + Haiku evaluation)
- Ingest — Convert to rich SQLite (reactions, roles, mentions)
- Generate — Create daily memory files
- Simulate — Process days chronologically to grow the soul
- Birth — Add to OpenClaw with config + binding
- Maintain — Cron job updates memory daily
Prerequisites
- OpenClaw — AI agent framework (GitHub | Docs)
- DiscordChatExporter CLI
- Python 3.10+
- Discord token in
~/.config/discord-exporter-token
Scripts
| Script | Purpose |
|---|---|
create_agent.sh | Full pipeline: Discord → Agent |
ingest_rich.py | JSON → SQLite with reactions/roles |
generate_daily_memory.py | SQLite → daily markdown |
simulate_growth.py | Generate soul emergence prompts |
incremental_export.sh | Fetch new messages only |
update_agent.sh | Daily cron wrapper |
regex-filter.py | Fast injection pattern detection |
evaluate-safety.py | Haiku semantic safety evaluation |
secure-pipeline.sh | Full security pipeline |
Security
⚠️ Discord content may contain prompt injection attacks.
Before ingesting content to your agent:
- Run
regex-filter.py— Fast pattern matching - Run
evaluate-safety.py— Haiku semantic evaluation - Only use messages with
safety_status = 'safe'
See SKILL.md for full security documentation.
Templates
Agent workspace files in templates/:
SOUL.md— Community identityMEMORY.md— Long-term milestonesLEARNINGS.md— Discovered patternsAGENTS.md— Key figuresTOOLS.md— Channels and ritualsHEARTBEAT.md— Maintenance protocol
See SKILL.md for the complete guide.
OpenClaw Integration
This skill is designed to work with OpenClaw, an open-source AI agent framework.
Key features used:
- Agent workspaces — Each Discord community becomes its own agent
- Memory search — Semantic search across daily memory files
- Heartbeats — Periodic maintenance to consolidate learnings
- Bindings — Connect agents to Telegram, Discord, Slack, etc.
Get started with OpenClaw:
- 🌐 Website: openclaw.ai
- 📚 Docs: docs.openclaw.ai
- 💻 GitHub: github.com/openclaw/openclaw
- 💬 Discord: discord.gg/clawd
License
MIT