submit-product-directories-v1-batch

审查 · 64
已收录

SPD V1 Batch. Process large, user-supplied sets of legitimate product, software, startup, app, and AI-tool directory URLs across Windows, macOS, and Linux-capable environments through normalization, deduplication, execution sharding, verification-first queues, authorization-contr

Verified installs0
Stars483
版本1.0.0
质量74/100 ·
信任64/100 · 仅限沙盒
审计79/100 · 需审查

供给资产档案

编程与开发 Agent

代码审查、仓库分析、测试、CI、GitHub、DevOps 与开发工作流 Skill。

浏览赛道

场景

GitHub automation

I need my agent to triage GitHub issues, review pull requests, and summarize repository changes.

适配 Agent

Claude Code + Cursor + Browser agents

适用于 Codex、Claude Code、Cursor、CLI 或自定义 Agent。

安装

就绪

npx skills add flaqai/backlink_skills --skill submit-product-directories-v1-batch

维护状态

新鲜

距上次推送 1 天

风险

需审查

Dependency or permission surface needs review

GitHub 质量

483

74/100 质量 · 72/100 信任

覆盖标签

编程GitHub automation设计与创意agent-skill

审查说明

Dependency or permission surface needs review · Permission surface may require sandboxing

Agent 采用评分卡

一眼查看信任、审计与安装准备度

这些分数综合公开仓库元数据、OpenAgentSkill 审查信号、维护新鲜度与安装准备度。它用于候选筛选,不替代人工审查。

质量

74

可靠的选择,值得加入生产工作流候选列表。

信任

仅限沙盒
64

有用但信任信号不足或混杂的候选项。在结果闭环证明任务匹配前,请保持在隔离工作区内使用。

审计

需审查
79

对安装准备度、安全元数据、维护情况与采用风险的机器可读审查。

OpenAgentSkill 信任评分 v5

安装前需人工审查

仅在沙盒中运行,并在用于真实工作前比较接近的替代方案。

CodexClaude CodeCursorOpenAgentSkill CLI

Stars

483 个 GitHub Stars

仓库活跃度

483 个 Star,175 个 Fork

维护状态

距上次推送 1 天

许可证

MIT

安装

npx skills add flaqai/backlink_skills --skill submit-product-directories-v1-batch

安装安全性

标准软件包或运行时安装路径

权限范围

secrets or environment access, shell or command execution

Agent 结果

暂未有 Agent 结果数据

文档

Usable metadata, review docs

风险摘要

生产前审查

  • Quality score needs review
  • Permission surface needs review: secrets or environment access, shell or command execution
  • Dependency/runtime risk: command execution surface, credential or environment access
  • Permission surface: secrets or environment access, shell or command execution

安装准备度

安装路径可用

  • 安装路径可用
  • 仓库证据可用
  • 已声明许可证
  • 暂无 Agent 验证结果证据

Agent 可读元数据

这个 Skill 的机器可读决策数据。

使用此区块或内嵌 JSON 判断 Agent 是否应安装该 Skill、选择替代方案,或先请求人工审查。

打开 JSON

适用任务

  • GitHub automation 工作流
  • Claude Code 团队
  • builders willing to evaluate younger projects
  • Inspect repository metadata

适用 Agent

CodexClaude CodeCursorOpenAgentSkill CLIBrowser agentsCLI

安装决策

命令
npx skills add flaqai/backlink_skills --skill submit-product-directories-v1-batch
策略
阻止
人工审查

信任与风险

信任
64/100
审计
79/100
风险级别
需审查

结果闭环

端点
/api/agent/outcome
事件 ID
resolve
结果
5

安装命令

npx skills add flaqai/backlink_skills --skill submit-product-directories-v1-batch

不适用场景

  • 需要厂商支持 SLA 的团队
  • 没有内部安全审查的高合规环境
  • 当前元数据中未发现重大风险信号
  • 高风险权限提示:Shell or command execution, Secrets or environment access
  • Dependency or permission surface needs review

Agent 安全 v2

35/100 · 避免自动安装

Blocked for auto-install阻止

This skill should not be selected by an agent without explicit human security review.

Do not auto-install. Inspect the source, dependencies, and permission surface first.

通过 API 解析

Shell 或命令执行

Skill 元数据引用了终端、CLI、Shell、子进程或命令执行工作流。

Browser automation

Skill may drive a browser or interact with web pages.

网络访问

Skill 可能访问远程页面、API、仓库或外部服务。

Secrets or environment access

Skill metadata references credentials, tokens, environment variables, or secret-bearing workflows.

  • 高风险权限提示:Shell or command execution, Secrets or environment access
  • Dependency or permission surface needs review

安装目标

在你的 Agent 工作流中安装此 Skill

通过公开安装端点获取命令、安全清单、目标提示词和该 Skill 的规范链接。

skill install

OpenAgentSkill CLI

Resolve policy, run the source installer safely, and report a verified install receipt.

$ npx --yes https://github.com/Leon-Drq/openagentskill/releases/download/cli-v0.2.1/openagentskill-0.2.1.tgz install flaqai-submit-product-directories-v1-batch

Agent 解析计划

让 Agent 在安装前验证匹配度。

Resolve API 返回首选 Skill、替代方案、安全策略、审计说明、安装目标和可直接执行的提示词,无需抓取此页面。

打开文本计划

Agent 应检查

  • 从 Resolve API 检查任务匹配与替代方案。
  • 检查审计评分、信任评分和安全策略警告。
  • 检查 Codex、Claude Code、Cursor 或 CLI 的安装目标兼容性。

复制提示词

Task: Use submit-product-directories-v1-batch in this workspace.
Resolve first: https://www.openagentskill.com/api/agent/resolve?task=Use%20submit-product-directories-v1-batch%20for%20an%20agent%20workflow&agent=codex&max_risk=medium
Review install handoff: https://www.openagentskill.com/api/skills/flaqai-submit-product-directories-v1-batch/install
Install command: npx skills add flaqai/backlink_skills --skill submit-product-directories-v1-batch
Before running it, summarize audit warnings, required permissions, and the fallback skill if install is risky.

Agent 交接

把安装路径交给 Agent,而不是再给一个目录页。

通过公开安装端点获取命令、安全清单、目标提示词和该 Skill 的规范链接。

打开安装 API

Agent 提示词

Use submit-product-directories-v1-batch for this task. Review https://www.openagentskill.com/api/skills/flaqai-submit-product-directories-v1-batch/install, then install with: npx skills add flaqai/backlink_skills --skill submit-product-directories-v1-batch

Registry 元数据

用于自动选择 Skill 的 Agent 可读档案。

本页通过 Registry API 提供相同的决策、信任、审计、场景和安装信号,让 Agent 无需抓取界面即可排序。

打开 Manifest

适配 Agent

76/100

GitHub automation

平台

Claude Code, Cursor, Browser agents

审计报告

需审查 · 79/100

对安装准备度、安全元数据、维护情况与采用风险的机器可读审查。

查看审计报告查看评估报告

Agent 决策面板

Companion skill for GitHub automation

将此 Skill 加入候选列表,并在生产使用前与相近替代方案比较。

76
就绪度
候选列表
阶段

栈中角色

辅助 Skill

主要匹配

GitHub automation

信任标签

强候选

安装路径

命令已就绪

适用场景

  • GitHub automation 工作流
  • Claude Code 团队
  • builders willing to evaluate younger projects

证据

  • 仓库近期活跃
  • 已提供安装命令或 GitHub 仓库
  • 74/100 质量档案
  • 13 个 OpenAgentSkill 交互事件

先审查

  • 当前元数据中未发现重大风险信号

实施路径

  1. 1在沙盒 Agent 中安装它,并端到端完成一次GitHub automation任务。
  2. 2Compare output quality, latency, and failure behavior against at least one alternative.
  3. 3Promote it into production only after reviewing repository permissions, license, and maintenance signals.

信任档案

仅限沙盒

有用但信任信号不足或混杂的候选项。在结果闭环证明任务匹配前,请保持在隔离工作区内使用。

64
OpenAgentSkill 信任评分

GitHub 采用度

信息

483 个 GitHub Stars

Star/Fork 活跃度

信息

483 个 Star,175 个 Fork; 当前元数据中没有议题活跃度信息

近期维护

通过

距上次推送 1 天

许可证清晰度

通过

MIT

积极信号

  • AI 审查已通过
  • 安装路径可用
  • 仓库证据可用
  • 近期维护的仓库
  • 安装命令未发现明显高风险模式
  • 结果闭环已就绪,但需要首次真实 Agent 运行

安装前审查

  • Quality score needs review
  • Permission surface needs review: secrets or environment access, shell or command execution
  • Dependency/runtime risk: command execution surface, credential or environment access
  • Permission surface: secrets or environment access, shell or command execution
  • 暂未有真实 Agent 结果报告
  • 无人值守安装前需要人工审查

建议操作

仅在沙盒中运行,并在用于真实工作前比较接近的替代方案。

质量档案

适用于 Agent 工作流的候选

可靠的选择,值得加入生产工作流候选列表。

74
GitHub Stars
483
新鲜度
1 天前
安装就绪
许可证
MIT

工作流匹配

在这些场景使用此 Skill

工作流匹配

加入完整工作流

替代方案短名单

安装前对比

可能适合该任务的相近 Skill。

对比全部

概览

--- name: submit-product-directories-v1-batch description: SPD V1 Batch. Process large, user-supplied sets of legitimate product, software, startup, app, and AI-tool directory URLs across Windows, macOS, and Linux-capable environments through normalization, deduplication, execution sharding, verification-first queues, authorization-controlled form work, idempotent submission, recovery, and truthful throughput reporting. Use when coverage and operational throughput matter more than deep per-site quality analysis. Do not use for ranking manipulation, bulk link spam, invented data, CAPTCHA bypass, paid-link acquisition, forced reciprocal links, or routes prohibited by a site's terms. ---

# SPD V1 Batch — large-batch directory operations

## Version identity

- Canonical name: `SPD V1 Batch`. - Invocation: `$submit-product-directories-v1-batch`. - Optimize for queue throughput, repeatability, verification handling, and recovery across large source lists. - Apply a fast legitimacy gate, not the deeper editorial and referral-value analysis used by `$submit-product-directories-v2-quality`. - Route campaigns requiring careful site selection, durable-placement analysis, or SEO-quality evidence to V2 Quality.

## Load controls

1. Read the verified product profile, brand rules, contact and credential aliases, approved assets, source list, batch authorization, and existing record. 2. Read [references/workflow.md](references/workflow.md) before planning or browser work. 3. Read [references/status-model.md](references/status-model.md) before writing or auditing records. 4. Read [references/browser-control-routing.md](references/browser-control-routing.md) before any browser or app interaction. Run the Windows/macOS/Linux capability preflight and select the backend from the current environment; do not assume a specific browser, operating system, or Computer Use support. 5. Copy [assets/submission-record-template.md](assets/submission-record-template.md) when no V1 Batch record exists.

Never invent product, company, founder, pricing, address, launch, ownership, contact, or legal facts. Keep optional unknowns blank and block required unknowns.

## Apply the batch legitimacy gate

Reject or separate any route that is irrelevant to the product, unavailable, unreleased-only, paid-link-only, forced-reciprocal, a known low-quality directory network, or prohibited for automated form work. Do not select sites because they promise dofollow links, ranking gains, DA/DR, or backlink volume.

Use only the exact brand, product name, or naked canonical URL as public link text. Never request dofollow treatment or use repeated commercial exact-match anchors.

## Build the queue

1. Normalize hostnames and submission routes. Strip tracking parameters from the record while preserving required route parameters in controlled evidence. 2. Derive an idempotency key from platform domain, product canonical ID, account alias, and route. 3. Deduplicate before opening the browser. Never execute an idempotency key that is already submitted, awaiting approval, published, or outcome unknown. 4. Assign stable queue IDs and execution shards. Treat shard size and maximum active tabs as operational settings, not SEO safety thresholds. 5. Classify every site into `direct form`, `account required`, `manual verification`, `email verification`, `paid/reciprocal`, `unavailable`, `ineligible`, or `unknown`. 6. Use batch-scoped authorization only when it names the allowed actions, source-list scope, approver alias, approval time, and expiry. Payments, reciprocal-site changes, DNS changes, and publication outside a directory require separate authorization.

## Run the verification-first pipeline

1. Run a read-only preflight over each shard before entering product-listing fields. 2. Expose the earliest native CAPTCHA, Turnstile, image code, email check, login, or similar safeguard. 3. Attempt only the site's ordinary native automatic verification. Never bypass, outsource, or weaken a safeguard. 4. Move unresolved items to one manual queue and continue processing eligible sites. 5. After the user completes the queue, recheck token validity and process short-lived tokens first. 6. Do not hold more active challenge tabs than the configured browser capacity.

## Execute forms at scale

1. Process only sites that passed the legitimacy gate, authorization check, duplicate check, and verification prerequisite. 2. Reuse approved field variants by length and category, while preserving exact public brand spelling and truthful meaning. 3. Keep newsletters and optional promotions off unless authorized. 4. Review plan, cost, URL, identity, category, agreements, uploads, and verification immediately before submission. 5. Submit sequentially within a browser profile. Record the result before advancing the queue cursor. 6. Never retry an ambiguous final action. Check the account backend, mailbox, and public page first. 7. Save drafts, transient failures, manual actions, and terminal outcomes as distinct states so the campaign can resume without replaying completed work.

## Protect records

- Store aliases and controlled evidence IDs, not passwords, OTPs, recovery codes, cookies, OAuth parameters, magic links, raw session IDs, raw email addresses, phone numbers, or tokenized URLs. - Separate the shareable campaign record from controlled evidence. - Treat a click, registration, draft, cleared form, or generic thank-you URL as insufficient submission evidence.

## Close and measure

Run:

On macOS or Linux:

```bash python3 scripts/audit_submission_record.py path/to/v1-batch-record.md python3 scripts/audit_submission_record.py path/to/v1-batch-record.md --json ```

On Windows, use `py -3` or an equivalent Python 3 launcher:

```powershell py -3 scripts/audit_submission_record.py path\to\v1-batch-record.md py -3 scripts/audit_submission_record.py path\to\v1-batch-record.md --json ```

Report totals by queue state, verification state, shard, and outcome. Measure queue completion rate, verified submissions per operator hour, duplicate avoidance, recovery rate, and unresolved manual workload. Report published listings separately from submitted forms. Do not report submission volume as proof of SEO value.

## Bundled resources

- [references/workflow.md](references/workflow.md): sharding, verification queues, execution, and recovery. - [references/status-model.md](references/status-model.md): record schema and state invariants. - [references/browser-control-routing.md](references/browser-control-routing.md): backend-neutral browser selection, interaction, confirmation, recovery, and evidence rules. - [assets/submission-record-template.md](assets/submission-record-template.md): privacy-safe V1 Batch template. - `scripts/audit_submission_record.py`: batch integrity, secret, duplicate, and state auditor.

技术详情

版本
1.0.0
许可证
MIT
最近更新
2026年8月21日
发布时间
2026年8月21日

决策摘要

辅助 Skill

76
就绪
候选列表
阶段

仓库近期活跃

审计

安装审查

安装与采用审查

79
需审查
安全性
76/100
维护状态
100/100
安装
92/100
打开完整审计查看评估报告

Agent 验证证据

Agent 验证证据

来自解析、审查、安装和一次小范围运行后的结果报告。

0
已验证
Needs first agent run自动安装: 先审查最近: 未知
成功率
近期失败
结果
0
输出质量
失败
0
不相关
0
安装次数
0
风险拦截
0
需要配置
0
生产环境
0

暂时没有 Agent 结果数据。首次 Agent 执行可以通过 /api/agent/outcome 报告成功、需要设置、风险拦截、失败或不相关。

安装

加入 Agent 工作流

免费且开源. 在生产 Agent 中安装前请先审查报告。

增长闭环

分享工具包

X

为 submit-product-directories-v1-batch 准备的场景化草稿,可手动发布到 X。

策展说明
submit-product-directories-v1-batch: SPD V1 Batch. Process large, user-supplied sets of legitimate product, software, startup, app...

483 stars

https://www.openagentskill.com/skills/flaqai-submit-product-directories-v1-batch?ref=x
打开 X 草稿
可选:带安装命令的回复
Listing + install path for submit-product-directories-v1-batch:
https://www.openagentskill.com/skills/flaqai-submit-product-directories-v1-batch?ref=x

Install: npx skills add flaqai/backlink_skills --skill submit-product-directories-v1-batch
打开回复草稿

收录来源

Registry 收录

可认领

此列表来自公开来源,维护者认领获批前不会标记为官方。

创作者
flaqai
收录方
OpenAgentSkill 社区索引

归属链接指向公开仓库或创作者主页。创作者可认领列表以更新所有权信号。

认领此 Skill

所有者认领

认领此 Skill 页面

这条 Registry 收录 列表归属于 flaqai,但尚未标记为官方。认领后可增加已验证所有者信号,使后续发布、安装和审计更新更值得信赖。

创作者外链工具包

将证据徽章加入你的 README

在开发者评估仓库的位置展示规范页面、当前信任与审计信号,以及真实的 Agent 验证证据。

[![Listed on OpenAgentSkill](https://www.openagentskill.com/api/badge/flaqai-submit-product-directories-v1-batch?metric=listed&label=Listed)](https://www.openagentskill.com/skills/flaqai-submit-product-directories-v1-batch)
[![OpenAgentSkill Trust](https://www.openagentskill.com/api/badge/flaqai-submit-product-directories-v1-batch?metric=trust&label=Trust)](https://www.openagentskill.com/skills/flaqai-submit-product-directories-v1-batch)
[![OpenAgentSkill Audit](https://www.openagentskill.com/api/badge/flaqai-submit-product-directories-v1-batch?metric=audit&label=Audit)](https://www.openagentskill.com/skills/flaqai-submit-product-directories-v1-batch/audit)
[![Agent Proven](https://www.openagentskill.com/api/badge/flaqai-submit-product-directories-v1-batch?metric=proven&label=Agent%20Proven)](https://www.openagentskill.com/skills/flaqai-submit-product-directories-v1-batch)

作者

F

flaqai

@flaqai

健康信号

GitHub Stars
483
质量评分
42/100
最近 GitHub 推送
2026年8月21日
框架提示
未知
OpenAgentSkill 浏览量
13
复制安装命令
0
跳转点击
0

社区信号

告诉我们这个 Skill 是否对你的 Agent 工作流有帮助。汇总反馈会持续改善排序。

信任与安全

仅限沙盒

64
  • GitHub 采用度483 个 GitHub Stars信息
  • Star/Fork 活跃度483 个 Star,175 个 Fork; 当前元数据中没有议题活跃度信息信息
  • 近期维护距上次推送 1 天通过
  • 许可证清晰度MIT通过
  • README/SKILL.md 完整度公开元数据需要更完整的 README/SKILL.md 上下文信息
  • 依赖与运行时风险command execution surface, credential or environment access检查