MODELROUTE

Local model router / proxy across Ollama, vLLM, and cloud with fallback

AI Agents & LLMOps — build, route, evaluate, and secure agents.

pip install cognis-modelroute
modelroute scan .            # → prioritized findings in seconds

🔎 Example output

Real, reproducible output from the tool — runs offline:

$ modelroute-emit --version
modelroute 0.1.0

$ modelroute-emit --help
usage: modelroute [-h] [--version] [--format {table,json}]
                  {route,simulate,providers,models} ...

Local model router/proxy with fallback.

positional arguments:
  {route,simulate,providers,models}
    route               resolve alias to a fallback chain + request plan
    simulate            route + dispatch with simulated outages
    providers           list configured providers
    models              list models (optionally filter by alias)

options:
  -h, --help            show this help message and exit
  --version             show program's version number and exit
  --format {table,json}

Blocks above are real modelroute output — reproduce them from a clone.

Sample result format (illustrative values — run on your own data for real findings):

{
"finding": {
"id": "1234567890",
"name": "Suspicious Network Traffic",
"description": "Network traffic from unknown IP address",
"confidence": 0.8,
"created_by": "AI System",
"created_at": "2023-02-20T14:30:00Z"
},
"indicators": [
{
"type": "ip",
"value": "192.168.1.100",
"label": "Malicious IP Address"
}
]
}

Related MCP server: promptpack

Usage — step by step

modelroute is a local model router/proxy that resolves a model alias into a provider fallback chain and builds the dispatch request. Console script: modelroute.

1. Install from a clone:

   pip install -e .

2. Resolve an alias into a fallback chain + request plan:

   modelroute route fast --prompt "Summarize this changelog" --strategy local-first

3. Inspect what's configured — list providers and models:

   modelroute providers
   modelroute models fast

4. Read the output — --format json returns the chosen candidate and full chain:

   modelroute --format json route fast -p "hi" | jq '.chosen, .fallback_chain'

5. Simulate an outage — verify failover by failing named providers:

   modelroute simulate fast -p "hi" --fail openai,anthropic

Contents

• Why modelroute? · Features · Quick start · Example · Architecture · AI stack · How it compares · Integrations · Install anywhere · Related · Contributing

Why modelroute?

AI infra

modelroute is single-purpose, scriptable, and self-hostable: point it at a target, get prioritized results in the format your workflow already speaks (table · JSON · SARIF), gate CI on it, and let agents drive it over MCP.

Features

• ✅ Resolve

• ✅ Build Request

• ✅ Estimate Tokens

• ✅ Messages Tokens

• ✅ Dispatch

• ✅ List Models

• ✅ List Providers

• ✅ Runs on Linux/macOS/Windows · Docker · devcontainer

• ✅ Ports in Python, JavaScript, Go, and Rust (ports/)

Quick start

pip install cognis-modelroute
modelroute --version
modelroute scan .                       # scan current project
modelroute scan . --format json         # machine-readable
modelroute scan . --fail-on high        # CI gate (non-zero exit)

Example

$ modelroute scan .
  [HIGH    ] MOD-001  example finding             (./src/app.py)
  [MEDIUM  ] MOD-002  another signal              (./config.yaml)

  2 findings · risk score 5 · 38ms

Architecture

flowchart LR
  IN[target / manifest] --> P[modelroute<br/>checks + rules]
  P --> OUT[findings (JSON / SARIF)]

Use it from any AI stack

modelroute is interoperable with every popular way of using AI:

• MCP server — modelroute mcp (Claude Desktop, Cursor, Cognis.Studio, uncensored-fleet)

• OpenAI-compatible / JSON — pipe modelroute scan . --format json into any agent or LLM

• LangChain · CrewAI · AutoGen · LlamaIndex — wrap the CLI/JSON as a tool in one line

• CI / scripts — exit codes + SARIF for non-AI pipelines

How it compares

Built in the spirit of LiteLLM, re-framed the Cognis way. Missing a credit? Open a PR.

Integrations

Pipes into your stack: SARIF for code-scanning, JSON for anything, an MCP server (modelroute mcp) for AI agents, and a webhook forwarder for SIEM/Slack/Jira. See docs/INTEGRATIONS.md.

Install — every way, every platform

pip install "git+https://github.com/cognis-digital/modelroute.git"    # pip (works today)
pipx install "git+https://github.com/cognis-digital/modelroute.git"   # isolated CLI
uv tool install "git+https://github.com/cognis-digital/modelroute.git" # uv
pip install cognis-modelroute                                          # PyPI (when published)
docker run --rm ghcr.io/cognis-digital/modelroute:latest --help        # Docker
brew install cognis-digital/tap/modelroute                             # Homebrew tap
curl -fsSL https://raw.githubusercontent.com/cognis-digital/modelroute/main/install.sh | sh

Related Cognis tools

• agentsmith — Config-first scaffolding and orchestration for multi-agent workflows

• skillhub — Local skill registry and installer for AI agents

• toolguard — Runtime allowlist and policy for agent tool-calls

• evalbench — Offline LLM / agent eval harness with regression gates

• ragkit — Batteries-included local RAG pipeline — ingest, index, serve

• memorybank — Portable long-term memory store for agents, exposed over MCP

Explore the suite → 🗂️ all 170+ tools · ⭐ awesome-cognis · 🔗 cognis-sources · 🤖 uncensored-fleet · 🧠 engram

Contributing

PRs, new rules, and demo scenarios are welcome under the collaboration-pull model — see CONTRIBUTING.md and SECURITY.md.

⭐ If modelroute saved you time, star it — it genuinely helps others find it.

Interoperability

{} composes with the 300+ tool Cognis suite — JSON in/out and a shared OpenAI-compatible /v1 backbone. See INTEROP.md for the suite map, composition patterns, and reference stacks.

License

Source-available under the Cognis Open Collaboration License (COCL) v1.0 — free for personal, internal-evaluation, research, and educational use; commercial / production use requires a license (licensing@cognis.digital). See LICENSE.