jebel-quant/rhiza-claude/status

Claude Code and agent skill by jebel-quant/rhiza-claude

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Report this repo's rhiza state — whether the config is valid, and what was actually synced. Read-only. Add --files for a tree of managed files, or --check for upstream drift.

Install

npx vskill@latest install jebel-quant/rhiza-claude/status

SKILL.md

You are running /status in the current working directory's repo.

A repo's rhiza state has two halves, and this reports both:

halffilequestion
config.rhiza/template.yml + repo structureis what we'd sync from well-formed?
sync.rhiza/template.lockwhat was actually synced, and when?

They can disagree in both directions, which is why reporting one without the other misleads: a freshly /init-ed repo has a valid config and no lock at all, and a long-synced repo can have a lock alongside a config someone has since broken by hand.

This command is a thin wrapper around two bundled scripts — plugin/scripts/validate.py and plugin/scripts/status.py, both deterministic and stdlib-only (no rhiza CLI, no PyYAML). Do not re-implement the parsing or gather fields yourself; run them and relay the output.

Purely descriptive: it reports; it does not score, fix, or file anything.

Argument (optional): $ARGUMENTS — a path to the repo root to inspect; default is the current directory.

1. Validate the config

${CLAUDE_PLUGIN_ROOT} resolves at runtime (keep the quotes); in a source checkout it's empty, so fall back to the repo-relative path.

bash
uv run --python 3.12 --no-project python "${CLAUDE_PLUGIN_ROOT}/scripts/validate.py" $ARGUMENTS

It checks more than the file: that the target is a git repo, that it has the expected language-specific structure (a pyproject.toml is required for Python), that template.yml exists and parses, and that its fields are present and well-typed. It exits non-zero on any of those failing — that's its contract as a gate, so it stays usable in CI. Here, treat a failure as a finding to report, not a reason to stop: still run step 2, because a broken config alongside a good lock is exactly the situation worth surfacing. Lead the report with the validation failure.

If there's no template.yml at all, the repo isn't rhiza-managed. Say so and point at /rhiza:init; skip step 2.

2. Report the sync state

bash
uv run --python 3.12 --no-project python "${CLAUDE_PLUGIN_ROOT}/scripts/status.py" $ARGUMENTS

Flags, added only when the user asks for them:

  • --json — machine-readable; the default is the human summary. (Its files array

is always present, so --json and --files don't combine usefully.)

  • --files (alias --tree) — list the managed files as a directory tree beneath the

summary. This is the view the retired /rhiza:tree gave.

  • --check — compare the pinned ref to the latest upstream release via

git ls-remote --tags (no gh, no auth) and print an Update line, e.g. v1.0.0 → v1.2.0 (2 releases behind) — run /update, or up to date. The only part of this command that touches the network; a git or network failure is reported, never fatal.

3. If a script can't run

If uv is missing, or a script isn't found at either path, report that plainly and stop — don't hand-roll the status or the validation as a substitute.

4. Relay the results

Show each script's output as-is; both are already formatted. Then, at most one short line tying the two halves together — whether the config is sound and how fresh the sync looks. No scores, no recommendations.

Two non-error states worth naming explicitly rather than reporting as failures:

  • `No template.lock found` — the repo is configured but never synced. Point at

/update, which performs the first sync.

  • Config valid, lock present, `--check` says behind — normal; that's what

/update is for.

For an assessment rather than a report, point the user at /quality.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/jebel-quant/rhiza-claude/status/badge)](https://verified-skill.com/skills/jebel-quant/rhiza-claude/status)

Something wrong with this skill or its page? Tell us.