Fable Mode is an open-source control plane for AI coding agents.
It makes an agent deliberate, produce evidence, and survive adversarial review before it earns permission to write to your workspace. The gates are mechanical, not prompt advice: no timer, no proof, no write access.
launch-film.mp4
A session starts locked. Confidence does not unlock it.
>>> session = create_session('demo-refactor', budget='2 min')
state=INIT execution_locked=True can_execute_code=False
>>> session.unlock_execution('the plan looks fine, let me write code now')
PermissionError: HARD TIME-LOCK VIOLATION: Execution unlock rejected!
The immutable 2.0m authority budget has not elapsed yet
(Remaining: 1m 59s / 120.0s).
The same gates guard every phase: evidence receipts for claims, a five-vector red-team swarm for code, and a sealed record of what was verified.
pip install fable-engineOr install the MCP server in your editor:
These links configure Fable Engine for AI agents in VS Code Chat or Cursor. They do not install a standalone editor extension. Both use uvx, which downloads and runs the pinned PyPI release in an isolated environment.
Point your agent at the MCP server manually:
Python 3.10+, zero runtime dependencies. Published on PyPI as fable-engine.
- Think — Time-locked deliberation. The agent cannot write until the timer ends.
- Prove — Claims need real evidence (tool receipts, hashes, invariants).
- Attack — A red-team swarm tries to break the code.
- Write — Only then is the workspace unlocked.
- Not a claim of flawless code. It is a checkable workflow, not a guarantee.
- Not a bigger prompt. The locks are enforced by the engine, not by wording.
- Not a framework lock-in. It speaks MCP and runs beside your current agent.
- V1 → V2 migration
- V2 architecture
- System 3 (experimental)
- Stop AI coding agents from writing too early
Issues and pull requests are welcome. See CONTRIBUTING.md.
MIT License · Built by REX