STAR: Command for Claude Code

.agents/commands/star-auto.md

star-auto is a command for Claude Code from wanghao9610/STAR. It costs 0 tokens per session (2,637 once invoked), scanned A, original, MIT.

An autonomous STAR workflow command that works toward a stated goal by choosing and running the next suitable research or development task.

In plain words
What is it for?
Use it to pursue a research, coding, planning, or project goal through connected STAR tasks until the goal is reached or a defined stop condition applies.
Why use it?
It removes the need to manually select every next step in a multi-stage plan, while still following the workflow’s approval and stopping rules.

Command for Claude Code

Written for Claude Code: a Claude Code command (commands/*.md). Also seen: mentions subagents; installed under .agents/ (shared by several agents).

This is wanghao9610/STAR's own configuration. It tells Claude Code how to work on STAR itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything STAR configures →

Reuse

Borrowing it

Nothing to install: this file belongs to wanghao9610/STAR. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/wanghao9610/STAR/main/.agents/commands/star-auto.md
Clone the repo
git clone --depth 1 https://github.com/wanghao9610/STAR

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for star-auto

README.md
[![agentmods](https://agentmods.dev/badge/commands/wanghao9610/star/star-auto.svg)](https://agentmods.dev/commands/wanghao9610/star/star-auto)
Your own site
<a href="https://agentmods.dev/commands/wanghao9610/star/star-auto"><img src="https://agentmods.dev/badge/commands/wanghao9610/star/star-auto.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,637 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00000 $0.02637
Opus 5 $0.00000 $0.01319
Sonnet 5 $0.00000 $0.00527
Haiku 4.5 $0.00000 $0.00264

Measured today against content hash 071122d5933a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

star-auto scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.agents/commands/star-auto.md · 55 lines

How it starts

The opening of the file, as written. The whole thing — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Drive the workflow toward a goal

The user has typed star-auto — this file is what that invocation runs. Typing it is the auto grant of conventions §10.7: while this run pursues its goal, the skill each next action names may be started without further naming, the seven explicit-only ones included, and a prepared STOP-line command may be launched instead of handed back. At involve=low, that same explicit grant also covers the executor chain's routine local implementation, mechanical review fixes, and Git lifecycle described below; a started run otherwise behaves exactly as if the user had typed its name.

Invocation shape: star-auto <GOAL> [stop=<STOP-LINE>] [involve=<level>].

Parse the invocation

  • Strip involve=<level> first, taken from outside the stop text — before stop=, or as the invocation's final token; the same words inside the stop line are stop-line text, not the token (conventions §7.7). When nothing sets a level — no token, no INVOLVE in .env — this run resolves to low, not medium: an autonomous run takes the marked recommendation on its own judgment calls and logs each (§7.8). At low, the invocation itself pre-approves the executor's recommended plan, explicit-path staging and commits, execution-branch or worktree creation, behavior-preserving review fixes that delete nothing, a clean reviewed squash merge, and removal of a worktree only after its durable artifacts have moved out and git status --porcelain is empty. Those operations run without an authorization question and are logged. Review fixes that require deletion are skipped and routed rather than guessed. The merged execution branch is retained without asking, so unattended cleanup never needs a forced delete. This exception is carried to the executor and code reviewer as the internal auto=unattended token; it is not part of the user-facing invocation shape.
  • Then strip stop=: everything from that token to the end of what remains is the stop line, natural language in the user's own words — stop=any single training over 4 hours, or any paid API call, stop=用卡超过 2 张就停 (either language, or both at once). Given, it is the line this run must not cross: before launching a prepared command, state its expected cost and judge it against each condition — a command that crosses the line, or that cannot be judged against it, is printed and handed back exactly as §2 says. Absent, the run is uncapped: prepared STOP-line commands launch autonomously, the goal owning the run (§2's one exception). Either way every launch is logged in the decisions record with its stated cost. Apart from the clean worktree removal above, neither the grant nor any stop line covers deletion, overwrite, sudo, system or driver installation, history rewriting, forced Git, push, discarding unmerged work, or resolving a merge conflict — those stop and report instead of being guessed through.
  • What remains is the goal. No goal — ask for one; never infer a goal from the repository.

Read the full file on GitHub · 55 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. today Changed · +1 lines 071122d5933a
  2. 8d ago First seen · 54 lines · 0 tokens per session scan A 8cd96237929d

Subscribe to this mod's changes

star-auto is a command published in the GitHub repository wanghao9610/STAR (51 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,637 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.