healthcheck-mapper
Healthcheck mapper: inventory all health/readiness/liveness endpoints, map against service dependencies, LLM judges each as adequate/weak/missing. Read-only. Trigger: /healthcheck
SKILL.mdskills/operations/healthcheck-mapper/SKILL.md
GitHub/healthcheck
Most healthchecks return 200 even when the DB is down. This skill audits coverage and quality.
What this is for
- Generic
/healththat doesn't check actual dependencies - Missing readiness probes, inconsistent check coverage across services
- Liveness vs readiness confusion
- Read-only skill. No endpoint modification, no deployment changes.
PROTECTION RULE - never ~/.claude/
Read-only skill. Guard required if write mode added later.
What You Must Do When Invoked
During analysis, assign a confidence level to each finding: proven (confirmed by evidence), likely (strong signal, needs review), or suspected (weak signal).
If /healthcheck -help or /healthcheck -h (without further arguments)
is invoked: output the ## Usage section unchanged and stop.
Otherwise follow these steps in order, skipping none.
Step 1 - Help check
If invoked with -help or -h, output the ## Usage section unchanged and stop.
Step 2 - Clarify target
Clarify -ProjectDir. Get confirmation.
Step 3 - Scan
& "<SKILL_DIR>/scripts/healthcheck-scan.ps1" -ProjectDir "<path>"
Step 4 - Analysis
Read each health endpoint:
- Does it check all critical dependencies (DB, cache, external API)?
- Is there a false-sense-of-security endpoint (always 200)?
- Is liveness vs readiness properly separated?
- What failure modes are covered?
Step 5 - Write report
File healthcheck-report.md in current working directory:
- Summary - total endpoints, adequate, weak, missing.
- Endpoint table - per endpoint: path, type, checked deps, missing deps, assessment.
- Per-service recommendations - what each endpoint should check.
- Open questions.
Step 6 - Summarize
State report path, highlight weakest endpoints.
Usage
/healthcheck # interactive
/healthcheck <dir> # scan project
/healthcheck -help