atm-doctor

name	atm-doctor
description	Run ATM health diagnostics, present severity-sorted findings, and delegate deep remediation analysis to the atm-doctor agent when critical findings exist.

ATM Doctor

This skill runs atm doctor --json, formats results for operators, and conditionally delegates to the atm-doctor background agent for remediation analysis.

Execution

1. Run doctor JSON and capture exit code:

set -o pipefail
tmp_json="$(mktemp)"
if atm doctor --json >"$tmp_json"; then
  doctor_ec=0
else
  doctor_ec=$?
fi
cat "$tmp_json"
rm -f "$tmp_json"

2. Parse JSON output and render a severity-sorted findings table:

• order: critical -> warn -> info
• columns: severity, code, check, message

3. Behavior by exit code:

• 0: show clean status summary; do not spawn agent.
• 2: spawn atm-doctor background agent and pass:
  • team
  • exit_code
  • full doctor JSON
• 1 or other non-zero: report command/runtime failure and stop.

Agent Delegation

When exit code is 2, invoke:

{
  "description": "Analyze atm doctor findings and propose remediation runbook",
  "prompt": "Analyze the provided atm doctor JSON and provide actionable remediation guidance.",
  "subagent_type": "atm-doctor",
  "run_in_background": true
}

Input payload to the agent must be fenced JSON:

{
  "team": "<resolved-team>",
  "exit_code": 2,
  "doctor_json": { "summary": {}, "findings": [], "recommendations": [], "log_window": {} }
}

Output Expectations

• Always present findings as a concise severity-sorted table.
• If delegated, append a short section:
  • Agent Analysis: condensed remediation steps
  • Escalate: any explicit user-escalation reasons
• Never emit raw tool traces in final user output.