GitHub High-Star Optimizer v1.2.0 — automated external distribution
GitHub High-Star Optimizer v1.2.0
Distribution Mode
This release adds dry-run-first external distribution automation while preserving the project’s no-code-change and no-engagement-manipulation boundaries.
Added
- Standard-library campaign publisher with dry-run as the default.
- Explicit
--execute --confirm <campaign-id>gate for live mutations. - DEV/Forem, Hashnode, Mastodon, LinkedIn, Discord, Slack, Telegram, and approved generic-webhook adapters.
- Assisted copy/open queue for Product Hunt, Hacker News, Reddit, V2EX, Juejin, Zhihu, and other rule-sensitive forums.
- Browser-assisted prefill guidance that can reduce manual work to final review/submit while preserving exact-destination authorization.
- Credential references through environment-variable names only.
- HTTPS enforcement, redirect refusal, credential redaction, bounded transient retries, per-target continuation, and maximum-target cap.
- Persistent idempotency state that skips an already successful identical payload and blocks changed-content reuse by default.
- JSON and Markdown results plus a local HTML review queue.
- Reusable distribution manifest template and full automation reference.
- Distribution capability synchronized across all 15 README languages.
- Local behavior tests using simulated endpoints; no external platform was contacted during validation.
Boundaries
- No target-project source code, dependency, build, test, CI, runtime, API, storage, permission, or behavior changes.
- No automated Stars, votes, follows, reviews, comments, account creation, CAPTCHA bypass, or community-rule evasion.
- Live destination setup still requires the user’s authorized accounts, tokens, webhooks, and exact campaign approval.
中文摘要
v1.2.0 新增“外部分发模式”:默认干跑检查,明确确认活动 ID 后,才会对已经启用且获得授权的平台执行 API 或 Webhook 发布。
自动渠道包括 DEV、Hashnode、Mastodon、LinkedIn、Discord、Slack、Telegram 和通用 Webhook;Product Hunt、Hacker News、Reddit、V2EX、掘金、知乎等规则敏感平台使用辅助发布队列。
新增凭据脱敏、HTTPS 强制、防重定向、有限重试、幂等防重复、逐平台结果记录和模拟端点测试。仍然不修改目标项目代码,也不自动刷 Star、点赞、关注、评论或评价。