agentleFS
all-source-osall-source-os/all-source/.claude/skills/fly-status/SKILL.md

fly-status

Skill4 starsChanged yesterdayApache-2.0View on GitHub

Check the status of all AllSource services on Fly.io. Shows running state, health checks, deployed versions, recent crashes/errors, and staleness vs latest git tag. Triggers on "fly status", "check fly", "deployment status", "are services running", "check production", "fly report", "fly health".

  • Commits and pushes

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    All AllSource backend apps on Fly.

    line 12
  2. Very specific:

    If an allsource-web app ever appears in fly apps list, flag it immediately — it was destroyed on 2026-04-15 and should not be recreated.

    line 23
  3. Distinctive:

    If a machine is stopped, check why:

    line 63
  4. Distinctive:

    Check if the latest release workflow succeeded (Docker images built):

    line 83

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Fly.io Status Report

Generate a comprehensive status report for all AllSource services deployed on Fly.io.

Apps to Check

All AllSource backend apps on Fly. The web frontend is on Vercel at https://all-source.xyz — it is NOT on Fly and should never be added here.

App NameServiceExpected State
allsource-coreRust event storealways running
allsource-queryElixir query servicealways running
allsource-control-planeGo control planeauto-stop (may be stopped when idle)
allsource-authRust auth servicealways running
allsource-primeRust Prime MCP serveralways running
allsource-registryRust package registryalways running

If an allsource-web app ever appears in fly apps list, flag it immediately — it was destroyed on 2026-04-15 and should not be recreated. The frontend lives on Vercel.

Also check alphasigmapro-* apps if they exist.

Procedure

1. List all apps

fly apps list 2>&1 | grep -v "Warning"

2. Machine status for each app

For each app, run:

fly machines list -a <APP_NAME> 2>&1 | grep -v "Warning"

Capture: state, checks, region, image, last updated.

3. Recent errors/crashes

For each app, check logs for crashes:

fly logs -a <APP_NAME> --no-tail 2>&1 | grep -iE "error|crash|panic|SIGKILL|OOM|exit|restart|unhealthy" | tail -10

Important: fly logs can be slow. Set a 15-second timeout per app. If it hangs, skip and note "logs unavailable".

Distinguish normal from abnormal:

  • exit_code=0, requested_stop=true → auto-stop (normal for control-plane)
  • exit_code=0, oom_killed=false → clean shutdown (normal)
  • exit_code!=0 or oom_killed=true → crash, investigate
  • SIGKILL → OOM or Fly forced kill — investigate

4. Check machine events for stopped apps

If a machine is stopped, check why:

fly machines status <MACHINE_ID> -a <APP_NAME> 2>&1 | grep -E "State|exit|oom|Event"

5. Version staleness

Compare deployed images to the latest git tag:

# Latest tag
git describe --tags --abbrev=0

# For each app, extract deployment timestamp from fly status
fly status -a <APP_NAME> 2>&1 | grep "Image"

6. GitHub Actions status

Check if the latest release workflow succeeded (Docker images built):

gh run list --workflow=release.yml --limit 1 2>&1
gh run list --workflow=docker-build.yml --limit 1 2>&1

Output Format

Present results as a markdown table:

## Fly.io Deployment Report

### Service Status

| App | Status | Health | Last Deploy | Git Tag | Stale? |
|-----|--------|--------|-------------|---------|--------|
| allsource-core | Running | 1/1 | Mar 6 | v0.17.1 | YES (18d) |
| ... | ... | ... | ... | ... | ... |

### Crashes / Errors

[List any non-zero exits, OOM kills, or error-level log entries]
Or: "No crashes detected. All services healthy."

### Action Items

[List anything that needs attention: stale deployments, stopped services that shouldn't be, failed health checks]

Common Issues

SymptomCauseFix
Control plane stoppedAuto-stop on idleNormal — starts on next request
Health check warning on stopped machineMachine not runningNormal if auto-stop
Image very old (>7 days)No recent deployDeploy latest: fly deploy -a <app>
OOM killedMemory limit too lowScale up: fly scale memory 1024 -a <app>
Exit code 137SIGKILL (OOM or timeout)Check memory usage, increase limits

Deploy Commands (for reference only — do NOT auto-deploy)

# Core (needs monorepo root context)
fly deploy --config apps/core/fly.toml --dockerfile apps/core/Dockerfile

# Query Service
cd apps/query-service && fly deploy

# Control Plane
cd apps/control-plane && fly deploy

# Web — NOT ON FLY. Frontend is on Vercel at https://all-source.xyz.
# Do not run `fly deploy` for web. Redeploy via Vercel (git push to main, or `vercel --prod`).

# Prime (needs monorepo root context)
fly deploy --config apps/prime-mcp/fly.toml --dockerfile apps/prime-mcp/Dockerfile

# Registry
cd apps/registry && fly deploy

NEVER deploy without the user explicitly asking. This skill is read-only — report status only.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.