Aiden Decisions

Ask small questions locally. Build workflows from the answers. Measure when they are wrong.

Aiden Decisions wraps Ollama's native decision endpoint with typed questions, validated answers, caching, bounded decision graphs and optional MCP tools. The Python core has no third-party runtime dependencies.

It grew out of a local assistant project inspired by Jev's decision interface. It uses Together AI's Tev1 through Ollama; it does not contain Jev's weights or reproduce its undisclosed architecture. No model weights are bundled.

Try it

Install Python 3.11+ and Ollama with decision-model support. Tev1's model card requires Ollama 0.35 or later; a real decision request, rather than a health check alone, verifies compatibility.

git clone https://github.com/HumanMeatSack/aiden-decisions.git
cd aiden-decisions
python3 -m venv .venv
source .venv/bin/activate
python -m pip install .
ollama pull tev1:4b-q4_K_M
aiden-decisions serve

Open http://127.0.0.1:8841/ for the interactive demo. On Windows, activate .venv\Scripts\activate instead. If the port is occupied, use serve --port 8842.

For a first result without downloading a model:

aiden-decisions preflight "turn the living room TV on"

That example uses the bundled fictional intent catalog and makes zero model calls. It returns a draft intent; it does not connect to a TV.

Related MCP server: deepbom

What you get

flowchart LR
    A[Request or compact facts] --> B{Known example intent?}
    B -->|Yes| C[Deterministic draft]
    B -->|No| D[Local Ollama / Tev1]
    D --> E[Validate typed answers]
    E --> F[Cache and confidence evidence]
    C --> G[Return a suggestion or workflow result]
    F --> G

Ask several questions

from aiden_decisions import DecisionEngine

engine = DecisionEngine()
result = engine.decide({
    "state": "My card was charged twice. Please refund the extra payment.",
    "questions": {
        "queue": {
            "type": "choice",
            "instructions": "Choose the support queue.",
            "criteria": {"billing": "Payments or refunds", "other": "Other issues"},
        },
        "refund": {"type": "noul", "instructions": "Is a refund explicitly requested?"},
    },
})
print(result["answers"])

Run a decision graph

aiden-decisions workflow --file examples/refund_workflow.json
aiden-decisions decide --file examples/support_ticket.json

The example batches independent questions, applies a fixed policy and evaluates a conditional follow-up. It classifies a refund request; it cannot issue a refund.

Connect an MCP client

python -m pip install '.[mcp]'
aiden-decisions-mcp

Keep the HTTP service running in a separate terminal. Configure your client's stdio command to the installed aiden-decisions-mcp executable. The five tools are classify_or_score, route_request, decision_preflight, run_decision_workflow and decision_health.

Client configuration and optional Codex hooks · Architecture and API · Evaluation and limitations

Confidence means evidence, not certainty

A concentrated model distribution is not a measured chance of being right. Calibration is opt-in, tied to the exact model and ordered question map, and limited to its evaluated task. No personal calibration profile ships by default. Graphs cannot intercept an assistant's internal reasoning. The engine has no device drivers, credentials, shell executor or authority to carry out a suggestion.

The included synthetic corpora are reproducible diagnostics, not evidence of general accuracy or superiority over Jev. See the evaluation document for results from this exact standalone export and the failed gates that remain disabled.

Develop

python -m pip install '.[mcp]'
python -m unittest discover -s tests -v
python scripts/check_release.py
python scripts/benchmark_decisions.py

Unit tests use synthetic backends and need no model, hardware or cloud account. The benchmark uses local Ollama and does not operate devices.

The initial standalone release passed 65 unit checks and a separate 61-check core-only installation. Its fixed smoke benchmark matched 53/56 examples; these are small synthetic diagnostics. Full results and limitations include the misses and the calibration gates that stayed disabled.

Credits

• Ollama native decisions

• Tev1 model and limitations

• Together AI's training code

• TypeSafe's Jev interface, the inspiration

• Official Python MCP SDK

Independent project; no affiliation with these providers. Project code is MIT licensed. Separately downloaded models and dependencies retain their own terms.