Migration from OpenClaw

This guide covers migrating from OpenClaw (and other frameworks) to OpenFang. The migration engine handles config conversion, agent import, memory transfer, channel re-configuration, and skill scanning.

Quick Migration

Run a single command to migrate your entire OpenClaw workspace:

openfang migrate --from openclaw

This auto-detects your OpenClaw workspace at ~/.openclaw/ and imports everything into ~/.openfang/.

Options

# Specify a custom source directory
openfang migrate --from openclaw --source-dir /path/to/openclaw/workspace

# Dry run -- see what would be imported without making changes
openfang migrate --from openclaw --dry-run

Migration Report

After successful migration, a migration_report.md file is saved to ~/.openfang/ with a summary of everything that was imported, skipped, or needs manual attention.

Other Frameworks

LangChain and AutoGPT migration support is planned for a future release. Currently, only OpenClaw migration is supported.

What Gets Migrated

Item	Source (OpenClaw)	Destination (OpenFang)	Status
Config	`~/.openclaw/config.yaml`	`~/.openfang/config.toml`	✅ Fully automated
Agents	`~/.openclaw/agents/*/agent.yaml`	`~/.openfang/agents/*/agent.toml`	✅ Fully automated
Memory	`~/.openclaw/agents/*/MEMORY.md`	`~/.openfang/agents/*/imported_memory.md`	✅ Fully automated
Channels	`~/.openclaw/messaging/*.yaml`	`~/.openfang/channels_import.toml`	⚠️ Automated (manual merge)
Skills	`~/.openclaw/skills/`	Scanned and reported	⚠️ Manual reinstall
Sessions	`~/.openclaw/agents/*/sessions/`	Not migrated	Fresh start recommended
Workspace files	`~/.openclaw/agents/*/workspace/`	Not migrated	Copy manually if needed

Channel Import Note

Channel configurations (Telegram, Discord, Slack) are exported to a channels_import.toml file. You must manually merge the [channels] section into your ~/.openfang/config.toml.

Skills Note

OpenClaw skills (Node.js) are detected and listed in the migration report but not automatically converted. After migration, reinstall skills using:

openfang skill install <skill-name-or-path>

OpenFang automatically detects OpenClaw-format skills and converts them during installation.

Manual Migration Steps

If you prefer migrating by hand (or need to handle edge cases), follow these steps:

Initialize OpenFang

openfang init

Creates ~/.openfang/ with a default config.toml.

Convert Your Config

Translate your config.yaml to config.toml:OpenClaw (~/.openclaw/config.yaml):

provider: anthropic
model: claude-sonnet-4-20250514
api_key_env: ANTHROPIC_API_KEY
temperature: 0.7
memory:
  decay_rate: 0.05

OpenFang (~/.openfang/config.toml):

[default_model]
provider = "anthropic"
model = "claude-sonnet-4-20250514"
api_key_env = "ANTHROPIC_API_KEY"

[memory]
decay_rate = 0.05

[network]
listen_addr = "127.0.0.1:4200"

Convert Agent Manifests

Translate each agent.yaml to agent.toml:OpenClaw (~/.openclaw/agents/coder/agent.yaml):

name: coder
description: A coding assistant
provider: anthropic
model: claude-sonnet-4-20250514
tools:
  - read_file
  - write_file
  - execute_command
tags:
  - coding
  - dev

OpenFang (~/.openfang/agents/coder/agent.toml):

name = "coder"
version = "0.1.0"
description = "A coding assistant"
author = "openfang"
module = "builtin:chat"
tags = ["coding", "dev"]

[model]
provider = "anthropic"
model = "claude-sonnet-4-20250514"

[capabilities]
tools = ["file_read", "file_write", "shell_exec"]
memory_read = ["*"]
memory_write = ["self.*"]

Convert Channel Configs

OpenClaw (~/.openclaw/messaging/telegram.yaml):

type: telegram
bot_token_env: TELEGRAM_BOT_TOKEN
default_agent: coder
allowed_users:
  - "123456789"

OpenFang (add to ~/.openfang/config.toml):

[channels.telegram]
bot_token_env = "TELEGRAM_BOT_TOKEN"
default_agent = "coder"
allowed_users = ["123456789"]

Import Memory

Copy any MEMORY.md files:

cp ~/.openclaw/agents/coder/MEMORY.md \
   ~/.openfang/agents/coder/imported_memory.md

The kernel will ingest these on first boot.

Config Format Differences

Aspect	OpenClaw	OpenFang
Format	YAML	TOML
Config location	`~/.openclaw/config.yaml`	`~/.openfang/config.toml`
Agent definition	`agent.yaml`	`agent.toml`
Channel config	Separate files per channel	Unified in `config.toml`
Tool permissions	Implicit (tool list)	Capability-based (tools, memory, network, shell)
Model config	Flat (top-level fields)	Nested (`[model]` section)
Agent module	Implicit	Explicit (`module = "builtin:chat"`)
Scheduling	Not supported	Built-in (`[schedule]` section)
Resource quotas	Not supported	Built-in (`[resources]` section)
Networking	Not supported	OFP protocol (`[network]` section)

Tool Name Mapping

Tools were renamed between OpenClaw and OpenFang for consistency. The migration engine handles this automatically.

OpenClaw Tool	OpenFang Tool	Notes
`read_file`	`file_read`	Noun-first naming
`write_file`	`file_write`
`list_files`	`file_list`
`execute_command`	`shell_exec`	Capability-gated
`web_search`	`web_search`	Unchanged
`fetch_url`	`web_fetch`
`browser_navigate`	`browser_navigate`	Unchanged
`memory_search`	`memory_recall`
`memory_save`	`memory_store`
`sessions_send`	`agent_send`
`agent_message`	`agent_send`
`agents_list`	`agent_list`

New Tools in OpenFang

These tools have no OpenClaw equivalent:

Multi-Agent

agent_spawn - Spawn new agents
agent_kill - Terminate agents
agent_find - Search for agents

Task Management

task_post - Post task to board
task_claim - Claim available task
task_complete - Mark task complete
task_list - List tasks by status

Event System

event_publish - Publish custom events

Scheduling

schedule_create - Create scheduled job
schedule_list - List scheduled jobs
schedule_delete - Delete scheduled job

Vision

image_analyze - Analyze images

Location

location_get - Get location info

Provider Mapping

OpenClaw Name	OpenFang Name	API Key Env Var
`anthropic` / `claude`	`anthropic`	`ANTHROPIC_API_KEY`
`openai` / `gpt`	`openai`	`OPENAI_API_KEY`
`groq`	`groq`	`GROQ_API_KEY`
`ollama`	`ollama`	(none required)
`openrouter`	`openrouter`	`OPENROUTER_API_KEY`
`deepseek`	`deepseek`	`DEEPSEEK_API_KEY`
`together`	`together`	`TOGETHER_API_KEY`
`mistral`	`mistral`	`MISTRAL_API_KEY`
`fireworks`	`fireworks`	`FIREWORKS_API_KEY`

New Providers in OpenFang

vllm - Self-hosted vLLM inference server
lmstudio - LM Studio local models

Feature Comparison

Feature	OpenClaw	OpenFang
Language	Node.js / TypeScript	Rust
Config format	YAML	TOML
Multi-agent	Basic (message passing)	First-class (spawn, kill, find, workflows)
Agent scheduling	Manual	Built-in (reactive, continuous, periodic, proactive)
Memory	Markdown files	SQLite + KV store + semantic search + knowledge graph
Session management	JSONL files	SQLite with context window tracking
LLM providers	~5	11 (Anthropic, OpenAI, Groq, OpenRouter, DeepSeek, Together, Mistral, Fireworks, Ollama, vLLM, LM Studio)
Per-agent models	No	Yes (per-agent provider + model override)
Security	None	Capability-based (tools, memory, network, shell, agent spawn)
Resource quotas	None	Per-agent token/hour limits, memory limits, CPU time limits
Workflow engine	None	Built-in (sequential, fan-out, collect, conditional, loop)
Event triggers	None	Pattern-matching event triggers with templated prompts
WASM sandbox	None	Wasmtime-based sandboxed execution
Python runtime	None	Subprocess-based Python agent execution
Networking	None	OFP (OpenFang Protocol) peer-to-peer
API server	Basic REST	REST + WebSocket + SSE streaming
WebChat UI	Separate	Embedded in daemon
Channel adapters	Telegram, Discord	Telegram, Discord, Slack, WhatsApp, Signal, Matrix, Email
Skills/Plugins	npm packages	TOML + Python/WASM/Node.js, FangHub marketplace
CLI	Basic	Full CLI with daemon auto-detect, MCP server
MCP support	No	Built-in MCP server (stdio)
Process supervisor	None	Health monitoring, panic/restart tracking
Persistence	File-based	SQLite (agents survive restarts)

Troubleshooting

Migration reports “Source directory not found”

The migration engine looks for ~/.openclaw/ by default. If your OpenClaw workspace is elsewhere:

openfang migrate --from openclaw --source-dir /path/to/your/workspace

Agent fails to spawn after migration

Check the converted agent.toml for:

Valid tool names (see the Tool Name Mapping table)
A valid provider name (see the Provider Mapping table)
Correct module field (should be "builtin:chat" for standard LLM agents)

Skills not working

OpenClaw Node.js skills must be reinstalled:

openfang skill install /path/to/openclaw/skills/my-skill

The installer auto-detects OpenClaw format and converts the skill manifest.

Channel not connecting

After migration, channels are exported to channels_import.toml. You must merge them into your config.toml manually:

cat ~/.openfang/channels_import.toml
# Copy the [channels.*] sections into ~/.openfang/config.toml

Then restart the daemon:

openfang start

Next Steps

Creating Agents

Explore 30 pre-built agent templates

Skill Development

Reinstall and create custom skills

Workflows

Build multi-agent pipelines

Production

Deploy OpenFang in production

Get Started

Core Concepts

Autonomous Hands

Configuration

Integrations

Guides

Migration from OpenClaw

Quick Migration

Options

Migration Report

Other Frameworks

What Gets Migrated

Channel Import Note

Skills Note

Manual Migration Steps

Config Format Differences

Tool Name Mapping

New Tools in OpenFang

Multi-Agent

Task Management

Event System

Scheduling

Vision

Location

Provider Mapping

New Providers in OpenFang

Feature Comparison

Troubleshooting

Migration reports “Source directory not found”

Agent fails to spawn after migration

Skills not working

Channel not connecting

Next Steps

Creating Agents

Skill Development

Workflows

Production

Build docs developers (and LLMs) love

Get Started

Core Concepts

Autonomous Hands

Configuration

Integrations

Guides

Documentation Index

​Quick Migration

​Options

​Migration Report

​Other Frameworks

​What Gets Migrated

​Channel Import Note

​Skills Note

​Manual Migration Steps

​Config Format Differences

​Tool Name Mapping

​New Tools in OpenFang

Multi-Agent

Task Management

Event System

Scheduling

Vision

Location

​Provider Mapping

​New Providers in OpenFang

​Feature Comparison

​Troubleshooting

​Migration reports “Source directory not found”

​Agent fails to spawn after migration

​Skills not working

​Channel not connecting

​Next Steps

Creating Agents

Skill Development

Workflows

Production

Build docs developers (and LLMs) love

Quick Migration

Options

Migration Report

Other Frameworks

What Gets Migrated

Channel Import Note

Skills Note

Manual Migration Steps

Config Format Differences

Tool Name Mapping

New Tools in OpenFang

Provider Mapping

New Providers in OpenFang

Feature Comparison

Troubleshooting

Migration reports “Source directory not found”

Agent fails to spawn after migration

Skills not working

Channel not connecting

Next Steps