Jaime - Juju AI Medic Engine

Jaime

Jaime is a Juju machine subordinate charm that observes a related principal charm on the same host, collects bounded diagnostic context when it becomes unhealthy, and writes structured incident reports. It can optionally call an AI provider (Gemini or OpenRouter) to suggest a diagnosis. It observes and reports only; nothing is changed without explicit operator intent.

Full documentation: https://canonical-jaime-charm.readthedocs-hosted.com/en/latest/

Quickstart

# Deploy Jaime from CharmHub (0.1/stable channel)
juju deploy jaime --channel=0.1/stable

# Deploy a principal charm (e.g. postgresql) and relate it to Jaime
juju deploy postgresql --channel 16/stable
juju relate postgresql jaime

# Monitor
juju status

To update an existing deployment:

juju refresh jaime --channel=0.1/stable

Actions

juju run jaime/0 diagnose              # Basic principal info
juju run jaime/0 collect-context       # Write a context bundle and return its path
juju run jaime/0 generate-report       # Report for the current open incident
juju run jaime/0 get-suggestion        # AI suggestion for the current incident
juju run jaime/0 show-status           # Monitoring state for all units
juju run jaime/0 show-usage            # LLM token usage and cost per model
juju run jaime/0 list-incidents        # Incident list from the audit log
juju run jaime/0 reset                 # Clear all incidents and start fresh

Jaime defaults to observe mode. To enable AI-assisted suggestions, see the documentation above for configuring mode, provider, and api-token.

Local OpenAI-compatible providers

To use a local server that speaks the OpenAI Chat Completions protocol (e.g. Ollama) instead of a hosted provider, configure provider=remote with provider-remote-url and an exact model tag. The charm runs inside the LXD container, so provider-remote-url must name the host address as seen from the container, and Ollama must listen beyond localhost (OLLAMA_HOST=0.0.0.0):

juju config jaime mode=suggest provider=remote model=llama3.2 \
  provider-remote-url=http://<host-ip>:11434/v1

No api-token is needed for a token-less local server.