Misceo
Health Pass
- License — License: NOASSERTION
- Description — Repository has a description
- Active repo — Last push 0 days ago
- Community trust — 15 GitHub stars
Code Pass
- Code scan — Scanned 5 files during light audit, no dangerous patterns found
Permissions Pass
- Permissions — No dangerous permissions requested
No AI report is available for this listing yet.
Local Anthropic-compatible AI gateway with cheap-first routing, quality gates, safe model handoffs, and an embedded cost dashboard.
Misceo
English · 简体中文
[!IMPORTANT]
This is the official public documentation, issue tracker, and release home
for Misceo. The product is distributed as prebuilt binaries through npm.
Source code is not included in this repository.
Cut AI-agent costs without breaking conversations or tool loops.
Misceo is a local Anthropic-compatible gateway. It routes eligible requests
through a lower-cost model, checks the completed candidate, and escalates to a
stronger model when policy requires it.
Unlike a request-only router, Misceo also protects agent state: session
continuity, cross-model transcript compatibility, and ownership of active tool
loops.
Why Misceo
Strong-only routing is simple but pays strong-model cost for every request.
Cheap-only routing lowers spend but cannot recover when the answer is empty,
malformed, or inadequate.
A static router decides before generation. Misceo can also inspect the first
completed candidate before choosing which answer the client receives.
| Strategy | Decision point | Trade-off |
|---|---|---|
| Strong-only | Before generation | Consistent quality posture; strong-model cost on every request. |
| Cheap-only | Before generation | Lowest routing overhead; no automatic recovery. |
| Static router | Before generation | Uses request metadata but cannot inspect the generated answer. |
| Misceo | Before and after the first attempt | Accepts a checked candidate or regenerates with a stronger model. |
Misceo does not make a cheaper model equivalent to a stronger model. It makes
the cost-versus-quality policy explicit, observable, and configurable.
How a request flows
flowchart LR
A["Anthropic-compatible client"] --> B["Misceo local gateway"]
B --> C{"Protected route?"}
C -->|"Yes"| S["Strong backend"]
C -->|"No"| L["Lower-cost attempt"]
L --> G{"Structural gate"}
G -->|"Fail"| S
G -->|"Pass"| J{"Judge policy"}
J -->|"Accept or skip"| R["Return candidate"]
J -->|"Reject or fail closed"| S
S --> R2["Return regenerated answer"]
B --> O["Local history and dashboard"]
- Protected routes, open tool loops, active pins, and explicit rules run
before the cascade. - Eligible traffic first reaches the configured lower-cost backend.
- A structural gate rejects upstream failures, empty output, and malformed
tool calls. - Depending on the selected mode, an optional judge scores the visible
candidate against the latest user turn. - A passing candidate is returned. A rejected candidate is discarded and a
stronger backend generates the served answer. - Escalated conversations remain on the stronger backend for a configurable
number of fresh turns.
See Routing and handoffs for decision precedence,
session identity, tool-loop protection, and mode-specific failure behavior.
Agent-safe routing
Conversation continuity
Misceo groups requests by structured session identity when the client provides
it. Bootstrap, title-generation, and visible requests from one launch remain
part of the same conversation.
Cross-model handoff
Provider-specific thinking signatures, cache markers, model names, and message
invariants are handled at model-family boundaries. The original local audit
record remains available for inspection.
Tool-loop ownership
When a model opens a tool loop, that backend keeps ownership until the loop is
complete. Misceo does not move an active tool_use / tool_result exchange to
another model family halfway through.
First-visible-reply protection
For eligible structured sessions, known launch bootstrap traffic is separated
from the user's first visible request. The first visible reply uses the strong
backend before later requests enter normal routing.
Where data goes
The proxy, embedded dashboard, configuration, and traffic logs run locally.
Local does not mean offline: inference requests go to the providers selected by
your routing policy.
For an eligible cascade, the first provider receives the request. If the
candidate is rejected, the escalation provider also receives the request. A
configured judge receives the capped latest user turn, visible candidate, and
tool names.
Read Privacy and data before using Misceo with
sensitive prompts or exposing either listener beyond loopback.
Highlights
- One Anthropic-compatible endpoint for Claude, GLM, and custom backends.
- Three explicit postures:
quality-first,balanced, andsavings-first. - Structural validation plus an optional LLM judge.
- Session-aware model-family handoffs and tool-loop protection.
- Embedded dashboard with live traffic, history, cost, and deletion controls.
- Observable serving backend, judge result, latency, usage, and route reason.
- Manual backend handoff from the CLI or dashboard.
Requirements
- Node.js and npm for installation and the npm launcher.
- Provider credentials for the backends you enable.
- No separate Bun, Docker, or UI installation.
Published platform packages currently cover:
| Operating system | Architectures |
|---|---|
| macOS | arm64, x64 |
| Linux | glibc arm64, glibc x64 |
| Windows | x64 |
See Compatibility for client and platform boundaries.
Start in five minutes
Install the CLI:
npm install --global @misceo/cli
Do not install with --omit=optional; the platform executable is delivered as
an optional package.
Create a private configuration:
mkdir my-misceo
cd my-misceo
misceo init --mode balanced
Add the provider credentials you use to .env:
ANTHROPIC_API_KEY=your-anthropic-key
ZAI_API_KEY=your-zai-key
Validate and start:
misceo doctor
misceo start
doctor performs static checks. It does not contact providers, validate live
credentials, or prove model availability.
Connect Claude Code on macOS or Linux:
ANTHROPIC_BASE_URL=http://127.0.0.1:4141 \
ANTHROPIC_AUTH_TOKEN=misceo \
claude
For Windows PowerShell and a complete verification procedure, read
Getting started.
Default local addresses:
- Proxy:
http://127.0.0.1:4141 - Dashboard:
http://127.0.0.1:5141
Verify routing in 60 seconds
Send a normal prompt in Claude Code, followed by a second small request. In an
eligible structured session, the first visible response is protected and a
later request can exercise the cascade.
Open the dashboard and inspect Live traffic. Confirm the serving backend,
route or cascade note, status, latency, and judge score when a judge ran.
In another terminal:
curl http://127.0.0.1:4141/healthz
misceo status
A healthy response, a traffic row, and a visible session confirm that the
client is using Misceo instead of contacting a provider directly.
Routing modes
| Mode | Judge behavior | If the judge is unavailable |
|---|---|---|
quality-first |
Required | Escalate to the strong backend. |
balanced |
Used when configured | Serve a structurally valid lower-cost candidate. This is the default. |
savings-first |
Skipped | Continue from the structural gate without a judge call. |
Select a mode during initialization:
misceo init --mode quality-first
misceo init --mode balanced
misceo init --mode savings-first
Dashboard
The management UI is embedded in the installed executable. It includes:
- live routing state, sessions, handoffs, and pins;
- traffic detail with request, response, judge, status, and latency;
- token and cost fields with contextual explanations;
- searchable, paginated conversation history;
- deletion for one conversation or all recorded history.
The dashboard has no user login. Keep it on loopback unless trusted
authentication and network controls are placed in front of it.
Read Dashboard for listener options and data controls.
Evidence and limitations
Misceo does not publish a universal savings or quality percentage. Results vary
with tasks, models, providers, prices, region, and sampling behavior.
The public benchmark contract compares four policies on the same versioned
task set:
strong-onlycheap-onlycheap-first + structural gatecheap-first + judge
Reports must include task success, quality loss, cost, latency, escalation
rate, judge error rates, and handoff retention. No public result is included
until the corresponding task set and raw report can be published together.
CLI quick reference
| Command | Purpose |
|---|---|
misceo init |
Create a private starter .env. |
misceo doctor |
Perform static configuration checks. |
misceo start |
Start the proxy and embedded dashboard. |
misceo status |
Show current routing and session state. |
misceo handoff BACKEND |
Switch the active backend. |
misceo cost |
Report token cost and configured savings. |
misceo audit |
Run explicit paid compatibility probes. |
misceo logs |
Prune, reindex, or forget local history. |
Detailed flags and configuration examples are in
Getting started and
Configuration.
Documentation
| Guide | Purpose |
|---|---|
| Getting started | Install, configure, connect, verify, update, and uninstall. |
| Routing and handoffs | Cascade decisions, sessions, judges, and tool loops. |
| Configuration | Environment variables, endpoints, pricing, and listeners. |
| Compatibility | Supported platforms, clients, and protocol boundaries. |
| Dashboard | UI behavior, history, cost fields, and exposure safety. |
| Privacy and data | Provider flow, local logs, retention, and deletion. |
| Troubleshooting | Common installation, endpoint, port, and history problems. |
| Benchmark methodology | Reproducibility rules, metrics, and claim boundaries. |
Source availability, license, and support
This repository does not contain Misceo source code. Published binaries are
licensed under the Functional Source License 1.1 with the Apache 2.0 Future
License. Review the license before redistribution or competing use.
Use GitHub Issues for reproducible
bugs and documentation problems. Read SUPPORT.md before filing.
Report security vulnerabilities privately as described in
SECURITY.md.
Reviews (0)
Sign in to leave a review.
Leave a reviewNo results found