Follow-up to #191 (epic), UX Tier 1. Lets a user experiment with any image once, without editing config or registering an agent.
Scope
bbox run-image ghcr.io/me/aider-bbox:latest -- aider --model gpt-5
bbox run-image ghcr.io/me/aider-bbox:latest \
--name aider --env OPENAI_API_KEY --env 'AIDER_*' \
--memory 4g --cpus 4 --mcp --mcp-authz-profile safe-tools \
--egress-profile standard --allow-host api.openai.com:443 -- aider
This mode is explicitly ephemeral:
- no credential persistence by default
- no settings import by default
- no custom config-file injection unless requested
- no env forwarding unless explicitly requested
Notes / constraints
Acceptance criteria
bbox run-image IMAGE -- CMD boots the image, runs the command, forwards only explicitly requested env, and persists nothing.
Follow-up to #191 (epic), UX Tier 1. Lets a user experiment with any image once, without editing config or registering an agent.
Scope
bbox run-image ghcr.io/me/aider-bbox:latest -- aider --model gpt-5 bbox run-image ghcr.io/me/aider-bbox:latest \ --name aider --env OPENAI_API_KEY --env 'AIDER_*' \ --memory 4g --cpus 4 --mcp --mcp-authz-profile safe-tools \ --egress-profile standard --allow-host api.openai.com:443 -- aiderThis mode is explicitly ephemeral:
Notes / constraints
agent.Agentfrom flags via the existingconfig.AgentFromOverridepath (Add declarative bring-your-own agent support (#191 Phase 1) #200) and runs it through the normalSandboxRunner— no config-file mutation, no registry persistence.BBOX_*env vars (Add declarative bring-your-own agent support (#191 Phase 1) #200) are injected as for any agent.Acceptance criteria
bbox run-image IMAGE -- CMDboots the image, runs the command, forwards only explicitly requested env, and persists nothing.