
Mcloud Logs
- 671 installs
- 207 repo stars
- Updated July 31, 2026
- medusajs/medusa-agent-skills
mcloud-logs is an agent skill that executes mcloud logs and jq filtering against Medusa Cloud environments so developers can interpret backend and storefront runtime logs when debugging production commerce failures.
About
mcloud-logs in medusajs/medusa-agent-skills teaches developers to execute mcloud logs against Medusa Cloud environments when debugging checkout failures, webhook errors, or slow API paths in production commerce workloads. The skill wraps the Medusa Cloud CLI with Bash(mcloud logs*) and jq* for filtering backend or storefront logs by time range, deployment scope, and error text. Constraints document that --follow and --json are incompatible, so programmatic analysis should use bounded --from/--to windows with --json instead of streaming pipelines. Follow mode streams until Ctrl+C and must stay out of scripts. Default retrieval returns the last 500 log lines, giving agents a bounded starting corpus before deeper search. Developers reach for mcloud-logs when interpreting runtime backend or storefront logs on Medusa Cloud rather than guessing failure causes from local-only output. The workflow fits commerce engineers operating deployed Medusa backends who need jq-friendly JSON log slices or human-readable tail output during incident response on live storefront and API deployments.
- mcloud log retrieval
- Error trace filtering
- Production incident triage
- Commerce API diagnostics
- Runtime observability patterns
Mcloud Logs by the numbers
- 671 all-time installs (skills.sh)
- +73 installs in the week ending Aug 4, 2026 (Skillselion tracking)
- Ranked #226 of 1,435 DevOps & CI/CD skills by installs in the Skillselion catalog
- Data as of Aug 4, 2026 (Skillselion catalog sync)
npx skills add https://github.com/medusajs/medusa-agent-skills --skill mcloud-logsAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 671 |
|---|---|
| repo stars | ★ 207 |
| Last updated | July 31, 2026 |
| Repository | medusajs/medusa-agent-skills ↗ |
How do you query Medusa Cloud production logs?
Query and interpret Medusa Cloud runtime logs when debugging checkout failures, webhook errors, or slow API paths in production commerce workloads.
Who is it for?
Medusa commerce developers debugging production checkout, webhook, or API latency issues on Medusa Cloud backend or storefront deployments.
Skip if: Local-only Medusa development without Cloud deployments or teams that never use the mcloud CLI for runtime log access.
When should I use this skill?
User debugs Medusa Cloud checkout failures, webhook errors, slow API paths, or asks to stream or filter mcloud logs by deployment and time range.
What you get
Filtered mcloud log output, jq-parsed error slices, and time-bounded JSON log windows scoped to backend or storefront deployments.
- Filtered runtime log output
- jq-parsed error excerpts
- Time-bounded JSON log windows
By the numbers
- Default mcloud logs retrieval returns the last 500 log lines
- Allowed-tools scope Bash to mcloud logs* and jq* commands
Files
Cloud CLI: Logs Command
Execute mcloud logs to fetch runtime logs for a Cloud environment's backend or storefront.
Constraints
--followand--jsonare incompatible. For programmatic log analysis, use bounded time windows with--from/--toand--json.--followstreams until interrupted withCtrl+C— do not use in scripts or pipelines.- Default retrieves the last 500 log lines from the past 15 minutes.
Command
mcloud logs \
--organization <org-id> \
--project <project-id-or-handle> \
--environment <environment-handle> \
[options]Options
| Option | Description | Default |
|---|---|---|
-o/--organization <id> | Organization ID | Active context |
-p/--project <id-or-handle> | Project ID or handle | Active context |
-e/--environment <handle> | Environment handle | Active context |
-f/--follow | Stream logs continuously (incompatible with --json) | false |
--limit <1-5000> | Max log lines (non-follow mode only) | 500 |
--from <ISO8601> | Start of time range (e.g. 2026-04-22T10:00:00Z) | 15 minutes ago |
--to <ISO8601> | End of time range; if >15 min ago, must also pass --from | now |
--search <string> | Filter by substring (same as dashboard search bar) | — |
--deployment <id> | Filter by deployment or build ID | — |
--source <string> | Filter by source (repeatable) | — |
--metadata <key=value> | Filter by metadata field (repeatable; same key merges values) | — |
| `--type <backend\ | storefront>` | Log stream to query |
--json | Output as JSON (incompatible with --follow) | false |
Examples
# Basic log fetch (last 500 lines, last 15 min)
mcloud logs --json
# Search for errors
mcloud logs --search error --limit 1000 --json
# Filter for HTTP 500 errors via metadata
mcloud logs --metadata status=500 --limit 1000 --json
# Logs for a specific deployment (build or deployment ID)
mcloud logs --deployment bld_01ABC123 --json
# Structured output for agent analysis
mcloud logs --search error --json | jq '.[] | {timestamp, source, message}'
# Storefront logs
mcloud logs --type storefront --json
# Stream live logs (human-readable, not for scripts)
mcloud logs --follow
# Logs within a specific time range
mcloud logs --from 2026-04-22T10:00:00Z --to 2026-04-22T11:00:00Z --limit 1000 --json
# Logs from a time until now
mcloud logs --from 2026-04-22T10:00:00Z --json
# Multiple source filters
mcloud logs --source api --source worker --json
# Multiple metadata filters (HTTP 4xx and 5xx)
mcloud logs --metadata status=400 --metadata status=500 --limit 500 --jsonTime Range Notes
- Default window is the past 15 minutes.
- Pass
--fromwithout--toto fetch from a time until now. - Pass
--towithout--fromonly if--tois within the last 15 minutes; otherwise also pass--from. - Both
--fromand--toaccept ISO 8601 timestamps.
Related skills
How it compares
Pick mcloud-logs for Medusa Cloud runtime log CLI access; pick local Medusa server debugging skills when failures reproduce only on a developer machine without Cloud deployments.
FAQ
Can mcloud-logs use --follow and --json together?
mcloud-logs documents that mcloud logs --follow and --json are incompatible. For programmatic analysis, mcloud-logs uses bounded --from and --to windows with --json, reserving --follow for interactive streaming stopped with Ctrl+C rather than shell pipelines.
How many log lines does mcloud logs return by default?
mcloud-logs notes that mcloud logs default retrieval returns the last 500 log lines from a Medusa Cloud backend or storefront environment. Developers can narrow further with time ranges, deployment scope, and jq filters when investigating checkout or webhook failures.
When should developers invoke mcloud-logs?
mcloud-logs fits Medusa Cloud operators reading backend or storefront runtime logs, filtering by time range, searching errors, or scoping output to a specific deployment during production commerce incidents. Invoke it when local logs cannot explain checkout, webhook, or API laten