Frameworks
Pick the right framework
Hatcher runs three production agent frameworks — OpenClaw for skill-heavy automation, Hermes for adaptive long-running agents, and IronClaw for privacy-first Rust-native agents.
At a glance
OpenClaw
AdvancedSkill-heavy autonomous worker
Power users who need maximum capability: autonomous task execution, persistent workspace, skills, cron, and multi-channel integrations.
Hermes
IntermediateLearning agent
Agents that learn and adapt: long-running memory, live config editing, tool routing, and persistent research workflows.
IronClaw
IntermediateRust-native private runtime
Privacy-focused agents on a hardened Rust runtime: WASM-sandboxed extensions, private libSQL memory, per-tool permissions, and conversational automations.
Feature matrix
| Feature | OC OpenClaw | HM Hermes | IC IronClaw |
|---|---|---|---|
| 2,500+ ClawHub skills | |||
| Browser automation | |||
| Cron jobs & triggers | |||
| Multi-channel gateway | |||
| Persistent memory | |||
| Character personas | |||
| Plugin ecosystem | |||
| Social media native | |||
| Lightweight deploy |
Best for...
Not sure which framework to pick? Match your use case.
OpenClaw
AdvancedPower users who need maximum capability: autonomous task execution, persistent workspace, skills, cron, and multi-channel integrations.
Hermes
IntermediateAgents that learn and adapt: long-running memory, live config editing, tool routing, and persistent research workflows.
IronClaw
IntermediatePrivacy-focused agents on a hardened Rust runtime: WASM-sandboxed extensions, private libSQL memory, per-tool permissions, and conversational automations.
Performance benchmarks
Cold-start time and baseline memory usage on Hatcher shared infrastructure. Actual values vary by config.
Cold-start time
Lower is faster
Baseline memory usage
Lower is leaner
Configuration examples
Each framework has its own config format. Here's what a typical setup looks like.
# openclaw/config.yaml
model: deepseek/deepseek-v4-flash
provider: openrouter
skills:
- web_search
- calculator
- file_manager
- code_interpreter
platforms:
telegram:
token: "${TELEGRAM_BOT_TOKEN}"
discord:
token: "${DISCORD_BOT_TOKEN}"
memory:
backend: sqlite
path: ./workspace/memory.dbLLM support
Both frameworks support BYOK (Bring Your Own Key) for provider-paid usage. Hosted LLMs route through Hatcher-supported providers and meter usage with AI Credits.
OC OpenClaw
HM Hermes
IC IronClaw
* BYOK always free — provider-paid usage without Hatcher AI Credits. View BYOK setup guide →
Quick decision guide
I want maximum tools and automation
OC · OpenClaw
I want an agent that learns over time
HM · Hermes
I need multi-channel messaging and triggers
OC · OpenClaw
I need a memory-heavy research assistant
HM · Hermes
Ready to deploy?
Pick a framework, configure your agent, and go live in under 60 seconds. Free tier included.