OpenSkillIndex← back to search
claude-codeopenclaw

Cc-openclaw openclaw-status

Check OpenClaw gateway health, channel connectivity, agent status, and recent cron job results

install
source · Clone the upstream repo
git clone https://github.com/rahulsub-be/cc-openclaw
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/rahulsub-be/cc-openclaw "$T" && mkdir -p ~/.claude/skills && cp -r "$T/.claude/skills/openclaw-status" ~/.claude/skills/rahulsub-be-cc-openclaw-openclaw-status && rm -rf "$T"
OpenClaw · Install into ~/.openclaw/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/rahulsub-be/cc-openclaw "$T" && mkdir -p ~/.openclaw/skills && cp -r "$T/.claude/skills/openclaw-status" ~/.openclaw/skills/rahulsub-be-cc-openclaw-openclaw-status && rm -rf "$T"
manifest: .claude/skills/openclaw-status/SKILL.md
source content

OpenClaw Status Check

Check the health and status of the entire OpenClaw deployment.

Setup Detection

OPENCLAW_REPO=$(readlink ~/.openclaw/openclaw.json 2>/dev/null | sed 's|/.openclaw/openclaw.json||')

Detect deployment mode — multi-gateway or single-gateway:

TIER_CONFIGS=(~/.openclaw/configs/openclaw-*.json)
[[ -f "${TIER_CONFIGS[0]}" ]] && MULTI_GATEWAY=true || MULTI_GATEWAY=false

Steps — Multi-Gateway Mode

If 

MULTI_GATEWAY=true
, check each tier independently:

  1. Discover tiers and ports: For each config in 
    ~/.openclaw/configs/openclaw-*.json
    :
TIER=$(basename "$cfg" | sed 's/openclaw-//;s/.json//')
PORT=$(grep -o '"port": *[0-9]*' "$cfg" | grep -o '[0-9]*')
  1. Gateway health per tier:
curl -s http://localhost:$PORT/health
  1. LaunchAgent status per tier:
launchctl print gui/$(id -u)/ai.openclaw.gateway.$TIER 2>&1 | head -5
  1. Channel connectivity per tier — check each tier's log:
tail -50 ~/.openclaw-$TIER/gateway.log | grep -E "telegram|slack|whatsapp|starting provider|socket mode|error"
  1. WhatsApp restart loop — check any tier whose config contains 
    "whatsapp"
    :
tail -50 ~/.openclaw-$TIER/gateway.log | grep -c "auto-restart attempt"

If count > 3, WhatsApp is in a restart loop (needs re-auth: delete 

~/.openclaw/credentials/whatsapp/default/
and restart).

  1. Agent count per tier: Read each tier's config and count entries in 

    agents.list
    .

  2. Cron job status: Read 

    ~/.openclaw/cron/jobs.json
    and for each enabled job, report:

    • Job name
    • Agent ID (and which tier it belongs to)
    • Last run time and status
    • Consecutive errors (if any)
    • Next scheduled run

  3. Recent errors per tier:

tail -100 ~/.openclaw-$TIER/gateway.log | grep -i "error\|fail\|crash" | tail -10
  1. Present results as a per-tier summary table:
    • Each tier: name, port, running/stopped (PID), channel connectivity
    • Agents: total per tier, which have channel bindings
    • Cron: jobs with errors or that haven't run recently

Steps — Single-Gateway Mode

If 

MULTI_GATEWAY=false
, fall back to single-gateway checks:

  1. Read port from config:
PORT=$(grep -o '"port": *[0-9]*' ~/.openclaw/openclaw.json | grep -o '[0-9]*')
  1. Gateway health:
curl -s http://localhost:$PORT/health
  1. LaunchAgent status:
launchctl print gui/$(id -u)/ai.openclaw.gateway 2>&1 | head -5
  1. Channel connectivity:
tail -50 ~/.openclaw/logs/gateway.log | grep -E "telegram|slack|whatsapp|starting provider|socket mode|error"
  1. WhatsApp restart loop check:
tail -50 ~/.openclaw/logs/gateway.log | grep -c "auto-restart attempt"

  1. Agent count: Read 

    ~/.openclaw/openclaw.json
    and count entries in 
    agents.list
    .

  2. Cron job status: Read 

    ~/.openclaw/cron/jobs.json
    — same as multi-gateway.

  3. Recent errors:

tail -100 ~/.openclaw/logs/gateway.log | grep -i "error\|fail\|crash" | tail -10
  1. Present results in a clear summary table:
    • Gateway: running/stopped (PID, uptime)
    • Channels: which are connected, which are failing
    • Agents: total count, which have channel bindings
    • Cron: jobs with errors or that haven't run recently