| name | omero-runtime-verifier |
| description | Safe runtime-debugging workflow for OMERO Docker Extended, including Loki-first log triage, container venv discovery, and correct OMERO service-user usage. |
| origin | repo-local skill informed by ECC v1.10.0 workflow patterns |
OMERO Runtime Verifier
Use this skill for live runtime debugging, service-health checks, and container-local diagnostics.
Mandatory procedure
- Start with
AGENTS.md and the relevant troubleshooting doc.
- Use the Loki/Admin Tools path first for logs.
- Switch to container-network probes when host
localhost is not reachable from the agent environment.
- Resolve the active runtime virtualenv before Python import checks.
- Use the service account, not
root, for OMERO CLI commands.
Hard rules
- Never run OMERO CLI as
root inside omeroserver or omeroweb.
- Do not repeat a Docker-socket permission error as if it were a product failure.
- Do not trust host-shell
localhost probes from a sandboxed agent shell.
- Do not use plain container
python3 when the code is installed inside a virtualenv.
- Prefer
docker exec -i ... <<'EOF' patterns over heavily escaped nested heredocs.
Correct runtime patterns
docker exec -i <container> bash -s <<'SH'
...
SH
docker exec <container> bash -lc 'su <service-user> -s /bin/bash -c "HOME=/tmp <command>"'
Verification targets
- service health and container status
- runtime venv path and importability
- Loki-backed logs and diagnostics
- OMERO CLI connectivity using the correct user and flag ordering
- Celery worker/process startup behavior when relevant
Good outcome
Runtime triage follows the repo's documented procedure once, avoids invalid probes, and produces evidence that matches the real deployment architecture.