Install
$ agentstack add skill-952jo-thanks-star-thanks-star ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Thanks Star
Thank the verified upstream repository only after a download, clone, or install succeeds. Treat a GitHub star as an external account change, not as part of the download permission.
Requirements
- Use Python 3.9 or later.
- Use an authenticated GitHub CLI session (
gh auth status). - Resolve `
to the directory containing thisSKILL.md` before running the bundled script. - Never request, print, or store a GitHub token. Let
ghmanage authentication.
Workflow
- Preserve the exact source supplied to the successful download or install command. Prefer an
OWNER/REPOvalue, GitHub URL, installed Skill directory, or local Git checkout. - Confirm the acquisition succeeded. Do not star after search, preview, failed download, failed clone, or failed install.
- Immediately after installing this Skill, or at its first activation when same-turn invocation is unavailable, read the policy:
``text python /scripts/thanks_star.py --json policy show ``
- If
onboarding_requiredistrue, ask exactly one standing-authorization question before any GitHub lookup:
> Thanks Star 默认使用自动模式。是否确认今后在 Agent 成功获取开源项目后,自动为明确识别出的 GitHub 上游仓库点 Star?每次新增点赞都会提示您。
Handle the answer as follows:
- Explicit yes: run
policy set auto --authorize-account-changes, then continue with the preserved source. - Explicit no: run
policy set offand stop. - Explicit request for per-repository confirmation: run
policy set ask, then continue inaskmode. - No answer or ambiguous answer: stop. Do not resolve a repository or change GitHub.
- If the configured policy is
off, stop without resolving or starring anything. - Run the operation with the preserved source:
``text python /scripts/thanks_star.py --json thank ``
- Handle the returned
statusexactly:
initial_confirmation_required: follow step 4. This is a defensive result when onboarding was skipped; make no GitHub lookup first.confirmation_required: the user selectedask; ask, “已确认上游为OWNER/REPO。是否给该仓库点 Star?” Run again with--approveonly after an explicit yes.starred: always tell the user, “您已为OWNER/REPO点赞。” Localize when the conversation uses another language.already_starred: say it was already starred only when useful; never imply a new change.skipped_unverified: do not ask for approval and do not star. Briefly identify the unresolved source when relevant.disabled: make no account change and do not nag the user.would_star: report that this was a dry run; never imply an account change.
- In
askmode, after approval for this exact repository, run:
``text python /scripts/thanks_star.py --json thank --approve ``
Do not reuse one-repository approval for another repository. In configured auto mode, do not ask again; notify after every new star.
Policy modes
When no policy exists, default to auto with onboarding_required: true. This pending state is not authorization and performs zero GitHub lookup or mutation until the user confirms once.
auto: the default after one-time standing authorization; star verified upstream repositories without further questions and notify after every new star.ask: require explicit approval for each previously unstarred repository.off: perform no lookup and no star action.
Show or change the policy with:
python /scripts/thanks_star.py --json policy show
python /scripts/thanks_star.py --json policy set ask
python /scripts/thanks_star.py --json policy set off
After explicit first-time or later standing authorization, enable auto with:
python /scripts/thanks_star.py --json policy set auto --authorize-account-changes
Never add the authorization flag based only on a download, installation, generic automation request, or silence.
Source rules
- Accept only public repositories on
github.com. - Resolve GitHub-declared forks to the root source repository.
- Resolve a declared mirror only when its upstream is itself an unambiguous GitHub repository; otherwise skip it.
- Use installer-injected
metadata.github-repo,metadata.github-ref, andmetadata.github-tree-shatogether as Skill provenance. - Prefer a local Git
upstreamremote. If it exists but is not GitHub, skip instead of falling back to a GitHub mirror inorigin. - Otherwise inspect
origin, then verify it through the GitHub API. - Skip local archives without trusted provenance, non-GitHub sources, private repositories, deleted repositories, and malformed inputs.
- Never star every repository owned by an author or organization.
- Never star the fork or mirror merely because it was downloaded.
- Never guess an upstream from a package name, README text, search result, or repository contents.
- Never unstar a repository.
The GitHub API's fork and mirror metadata cannot identify an undeclared manual copy. If the available evidence cannot establish provenance, skip.
Read-only inspection
Inspect resolution without checking or changing star state:
python /scripts/thanks_star.py --json inspect
Preview the complete configured policy decision without changing GitHub:
python /scripts/thanks_star.py --json thank --dry-run
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: 952JO
- Source: 952JO/thanks-star
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.