| name | health-check-pipeline |
| description | Run a suite of health checks (bash commands, file existence, content matching) and produce a structured pass/fail report. Pure Python, no LLM dependency. |
Health Check Pipeline
Use this skill when a monitor agent needs to verify system state, validate artifacts, or confirm that previous steps produced the expected output.
Inputs
checks (required, list[dict]) — health check items.
type: "bash" | "file_exists" | "file_contains"
target: command string (for bash) or file path
expected (optional): expected content substring (for file_contains) or exit code (for bash, default 0)
output_path (required, str) — where to write the markdown report.
Outputs
{
"report_path": "/path/to/health_report.md",
"total": 5,
"passed": 4,
"failed": 1,
"details": [
{"check": "todo.py exists", "status": "passed", "elapsed_ms": 0.1},
{"check": "pytest passes", "status": "failed", "elapsed_ms": 2300, "error": "1 test failed"}
]
}
Check types
| Type | Target | Expected | Behaviour |
|---|
bash | shell command | exit code (default 0) | Runs command, checks return code |
file_exists | file path | — | Checks path.exists() |
file_contains | file path | substring | Reads file, checks if substring present |