fable-modeBLOCK
Open-source MCP control plane for AI coding agents: mechanical time-locks, evidence-gated proof receipts, red-team remediation, and persistent engineering memory.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Agents that think before they write.
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.
Demo
KERR // ORRERY
One self-contained HTML file. Raw WebGL, zero libraries, zero external assets, and zero build step.
https://github.com/user-attachments/assets/8287bbfe-e3ee-4dcf-ba0f-f9ff22ae79bd
7 renders rejected before final · 2 bugs caught · 10/10 red-team probes passed
Fable Mode overview
https://github.com/user-attachments/assets/27f4f8a2-b1bb-4398-a08c-bc9fd93d69d7
Quick start
A session starts locked. Confidence does not unlock it.
- Install the MCP server using one of the options below.
- Add the optional Agent Skill if you want the full workflow.
- Ask your agent to use Fable Mode for a concrete coding task and choose a time budget.
A new session starts with execution locked:
{
"action": "creaa2fcfacc4b9cOBSERVED · 2026-10-09Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add fable-engine -- uvx fable-engine==1.3.9
Trust audit
BLOCKgrade D · trust 68/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | FAIL |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (2 observation(s))
- Network
- declared (6 observation(s))
- Shell
- declared (2 observation(s))
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (18)
exec(compile(str(init.get("source", "")), "<fable-sandbox-target>", "exec"), namespace)issues.append(f"{path}:{node.lineno}: Prohibited call to eval()")if str(ip) == "169.254.169.254":
.coderabbit.yaml
data = f"Thread-{thread_id} iteration {j}: payload content {hashlib.md5(f'{thread_id}-{j}'.encode()).hexdigest()}"os.symlink("/etc/passwd", source / "fable_mode" / "__init__.py")"http://169.254.169.254/latest/meta-data/",
"http://127.0.0.1/secret",
"http://127.0.0.255/",
"http://10.0.0.1/internal",
"http://172.16.0.1/status",
"http://192.168.1.1/router",
self.assertEqual(store.get_bytes(ref, verify=False), b"trusted")
run._attestation_secret = bytes.fromhex(secret)
self.assertNotEqual(restored._red_team_receipt_key, bytes.fromhex("11" * 32))assets/launch-film.mp4
During the time-lock window (Phases 1, 2, and 3), the agent holds full privileges to interact with the environment and author artifacts:
Gates applied: no_behavioural_pass.
a2fcfacc4b9cfull audit observations/trust-audit/mcp-server/rex-codebase__fable-mode.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-09 | a2fcfacc4b9c | BLOCK | D | 68 | first audit |
Questions
What is the fable-mode MCP server?
Open-source MCP control plane for AI coding agents: mechanical time-locks, evidence-gated proof receipts, red-team remediation, and persistent engineering memory.
Is fable-mode safe to connect to an agent?
No — not without reading the findings first. The audit graded it D (68/100) and found 3 critical or high issues in the source. Each one is listed on this page with the file and line it is on.
What credentials does fable-mode need?
It reads FABLE_BROKER_WRITE_TOKEN_DIGEST, FABLE_BROKER_WRITE_TOKEN_DIGEST_FILE, FABLE_FORCE_UNLOCK_TOKEN and FABLE_TEST_SECRET from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does fable-mode run?
It speaks stdio, so it runs as a local process your client starts. It is published on PyPI as fable-engine.
How current is this page?
The grade is for one exact copy of the source (a2fcfacc4b9c), read on 2026-10-09. The repository is watched and re-audited when it changes.