hermit-doctorlisted

# Hermit Doctor Runs twelve read-only health checks against the current hermit install and surfaces the summary. Safe to run at any time. Produces no side effects beyond writing `.claude-code-hermit/state/doctor-report.json` and appending a summary block to SHELL.md. ## Steps 1. Run the check script: ```bash node ${CLAUDE_PLUGIN_ROOT}/scripts/doctor-check.js .claude-code-hermit ``` The script writes `.claude-code-hermit/state/doctor-report.json` and prints the same JSON to stdout. It exits 0 unconditionally — on any internal failure the failing check reports `status: "fail"` in its own entry rather than crashing the report. 2. Parse the JSON. For each of the eleven checks in the report (`config`, `hooks`, `state`, `cost`, `proposals`, `dependencies`, `permissions`, `docker-security`, `archive`, `reflect`, `scheduler`), emit one line using this format: - `✓ <id> — <detail>` when `status: ok` - `⚠ <id> — <detail>` when `status: warn` - `✗ <id> — <detail>` when `status: fail` 2.5. **Sandbox capability check** (twelfth check, run after step 2): Architectural note: this check is computed by the skill orchestrator, not by `doctor-check.js`. `state/doctor-report.json` therefore contains only the eleven checks emitted in step 2; the sandbox line is appended to the rendered summary and to SHELL.md but is not present in the JSON report. Tools that consume `doctor-report.json` programmatically should call `scripts/sandbox-probe.py` separately if th