Brane is different from prompt guardrails because Brane enforces policy at runtime when an agent attempts an action. Prompt guardrails influence model behavior. Brane controls whether the action is allowed to execute. Both can be useful, but they protect different boundaries.Documentation Index
Fetch the complete documentation index at: https://docs.brane.membranelabs.org/llms.txt
Use this file to discover all available pages before exploring further.
Comparison
| Control layer | What it checks | When it runs | What it prevents |
|---|---|---|---|
| Prompt guardrail | Instructions and model behavior | Before generation | Some unsafe reasoning or output |
| Output filter | Text returned by the model | After generation | Some unsafe text |
| Brane runtime policy | Tool calls and capabilities | Before or after action execution | Unsafe actions and capability use |
Why Runtime Policy Matters
If an agent sends an email, deletes a user, issues a refund, queries a database, calls an MCP tool, or writes memory, the consequence happens at the action boundary. Brane evaluates policy at that boundary:Use Brane With Existing Safety Layers
Brane does not replace:- Prompt engineering
- Model safety systems
- IAM
- Database permissions
- Network controls
- Secrets management
- Application authorization