nagi-cli 0.0.1

A reconciliation engine that declaratively defines the desired state of data and automatically converges toward that state

Nagi

Nagi monitors your data, detects drift from the desired state, and automatically runs the operation you defined to fix it.

Install

pip install nagi-cli

How It Works

Define the desired state of each asset and the sync to run when drift is detected. Then compile, evaluate, and sync.

# conditions: "clean_data.txt exists"
apiVersion: nagi.io/v1alpha1
kind: Conditions
metadata:
  name: clean-data-check
spec:
  - name: file-exists
    type: Command
    run: [test, -f, clean_data.txt]
---
# sync: "create the file"
apiVersion: nagi.io/v1alpha1
kind: Sync
metadata:
  name: build-clean-data
spec:
  run:
    type: Command
    args: [sh, -c, "echo '...' > clean_data.txt"]
---
# asset: "when drifted, run the sync"
apiVersion: nagi.io/v1alpha1
kind: Asset
metadata:
  name: clean-data
spec:
  onDrift:
    - conditions: clean-data-check
      sync: build-clean-data

nagi compile    # Validate and resolve resource definitions
nagi evaluate   # Evaluate conditions → reports "clean-data" as Drifted
nagi sync       # Run sync → creates clean_data.txt → re-evaluates as Ready
mv clean_data.txt dirty_data.txt
nagi evaluate   # Drifted again
nagi sync       # Restores clean_data.txt → Ready

nagi serve runs this evaluate-and-sync cycle continuously.

See the Quickstart for a full walkthrough, or browse the Documentation.

License

Apache License 2.0. See LICENSE for details.