Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Gingiris-1031/gingiris-skills --skill gingiris-seo-geo-agentgit clone --depth 1 https://github.com/Gingiris-1031/gingiris-skillsWrote 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.
[](https://agentmods.dev/skills/gingiris-1031/gingiris-skills/gingiris-seo-geo-agent)<a href="https://agentmods.dev/skills/gingiris-1031/gingiris-skills/gingiris-seo-geo-agent"><img src="https://agentmods.dev/badge/skills/gingiris-1031/gingiris-skills/gingiris-seo-geo-agent/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/gingiris-1031/gingiris-skills/gingiris-seo-geo-agent"><img src="https://agentmods.dev/badge/skills/gingiris-1031/gingiris-skills/gingiris-seo-geo-agent.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Data Exfiltration · line 378 Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
- medium MCP Rug Pull · line 640 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00364 | $0.07243 |
| Opus 5 | $0.00182 | $0.03622 |
| Sonnet 5 | $0.00073 | $0.01449 |
| Haiku 4.5 | $0.00036 | $0.00724 |
Grade A, and why
gingiris-seo-geo-agent 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 10d ago.
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.
How it starts
The opening of the file, as written. The whole thing — 661 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SEO/GEO Agent 运营 SOP v2
版本说明：这是 v2 草稿，基于 analook.com 和 gingiris.tools 的真实运营经验重写。 v1 给了你方法论框架，v2 给你的是「踩过坑之后该怎么跑」。
⚠️ 开始之前：为什么必须先收集 context
v1 版本最大的问题：agent 不知道你的站是新站还是老站、GSC 有没有接上、已经有哪些落地页——结果给出的策略全是通用建议，落不了地。
真实事故复盘：
- analook.com：GSC OAuth token 过期，agent 继续用旧缓存数据跑了 2 周，完全是在盲开车。排名在动，我们完全不知道。
- gingiris.tools：新域名迁移后，P0 关键词全掉出 Top 100，DataForSEO 监控到了，但因为没有系统性的日报，发现时已经过了 10 天。
- 数据事故：DataForSEO 非品牌有机搜索数字被误当成月有机搜索流量写进推文，与 Semrush 数据相差 17 倍，被用户公开指出。
这三件事告诉我：没有 context，运营就是在赌。
第一步：开局 Context 收集（对话式，必须完成）
Agent 第一次运行时，逐步引导用户提供信息，一次只问一个问题，不要一次倒出大表格。
用户回答后，agent 根据回答调整策略，再问下一个。全部收集完后，给出一份简短的「当前状态判断」再开始运营。
对话引导脚本（按顺序执行）
Q1（先问这个）
你的网站主域名是什么？大概什么时候建的？
→ 收到后判断：新站（< 6 个月）/ 成长站（6-24 个月）/ 老站（> 24 个月）
Q2
Google Search Console（GSC）接入了吗？
A) 已接入，最近有看数据 B) 已接入，但很久没看了（可能 token 过期） C) 还没接入
→ A：问他粘贴最近 7 天 Top 10 关键词（截图或文字）
→ B：提醒先验证 token，建议跑一下 gws auth login
→ C：停！本次第一任务就是接 GSC + GA4，其他全等
Q3
网站大概有多少月流量？权重怎么样？
A) 我有 Ahrefs / Semrush 数据，DR 大概是 ___ B) 我有 Similarweb / 其他估算，月流量大概 ___ C) 不知道，没有工具
→ A 或 B：记录数字 + 来源，提醒「如果只有一个来源，建议交叉验证」 → C：告知「没关系，我们从关键词角度入手，先用 KD ≤ 20 的词」
Q4
你最直接的竞品是谁？选或填都行：
A) 我知道，URL 是：___（填 2-3 个就够） B) 我不确定谁是竞品，你帮我判断
→ A：记录竞品 URL，用于后续竞品对比页策略 → B：根据用户的产品描述推荐 3 个最可能的竞品候选，让用户确认
Q5
网站现在有这些页面吗？（可多选）
☐ 首页 ☐ 定价页 ☐ 竞品对比页（如「XX alternative」） ☐ 博客 / 内容页 ☐ 其他落地页：___
→ 根据勾选情况判断缺什么，优先补竞品对比页和定价页
Q6（可选）
有没有最想排上去的关键词？比如你觉得用户搜什么词会找到你？
A) 有，我觉得是：___ B) 不确定，让你帮我找
→ A：记录为 P0 关键词，日报必须追踪 → B：根据竞品和产品描述推荐 3-5 个候选词，让用户选
全部收集后：输出状态判断
收集完后，agent 给出一段简短的判断，例如：
📊 当前状态：
- 网站：[域名]，建站约 [X] 个月，属于 [新站/成长站/老站]
- GSC：[已接入，最近有数据 / 未接入，第一任务是接入]
- 权重：DR [X]，策略关键词 KD 上限 ≤ [XX]
- 竞品：[A] / [B] / [C]
- 当前落地页：[N 页，缺 竞品对比页/定价页...]
- P0 关键词：[词1] / [词2] / [词3]
→ 接下来做什么：[Agent 给出 3 步以内的下一步]
判断矩阵（Agent 内部参考）
| 情况 | 判断 | 策略调整 |
|---|---|---|
| 建站 < 6 个月 | 新站 | 专注 KD ≤ 20 的长尾词，不碰主词，先跑页面覆盖率 |
| 建站 6-24 个月 | 成长站 | 可攻 KD 20-35，竞品对比页优先 |
| 建站 > 24 个月 | 老站 | 可以挑战 KD 35+ 的词，重点在 CTR 优化 |
| GSC 未接入 | 🚨 紧急 | 第一周唯一任务：接 GSC + GA4，其他都等 |
| DR < 10 | 极低权重 | 外链建设是 P0，内容只写竞品对比页 |
| DR 10-30 | 低权重 | 关键词严格控制 KD ≤ 35 |
| DR > 30 | 有一定权重 | 可以适当突破 KD 限制 |
What ships with it
5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 10d ago First seen · 661 lines · 364 tokens per session scan A 7c70bee8c0ca
gingiris-seo-geo-agent is a skill published in the GitHub repository Gingiris-1031/gingiris-skills (79 stars, last pushed 5d ago), licensed MIT. It adds 364 tokens to every session and 7,243 once invoked, about $0.0018 per session on Opus 5. 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.
Other skills, from other repositories
google-search-console
Skill "google-search-console" from VKirill/claude-lane-stack, covering version requirements (may 2026), usage, use this skill when, do not use this skill when and purpose.
citedy-seo-agent
Full-stack AI marketing toolkit — scout X/Twitter and Reddit for trending topics, discover and deep-analyze competitors, find content gaps, publish SEO- and GEO-optimized articles with AI illustrations and voice-over in 55 languages, create social media adaptations for X, LinkedIn, Facebook, Reddit, Threads…
dataforseo-brain
Scaffold and operate DataForSEO Brain, a source-cited Obsidian brain for DataForSEO v3 API mastery - 12 modules and 200+ endpoints (SERP, Keywords Data, AI Optimization, Domain Analytics, DataForSEO Labs, OnPage, Backlinks, Content Analysis, Merchant, App Data, Business Data, Databases), the task/live queue execution…
citedy-seo-agent
Full-stack AI marketing toolkit — scout X/Twitter and Reddit for trending topics, discover and deep-analyze competitors, find content gaps, publish SEO- and GEO-optimized articles with AI illustrations and voice-over in 55 languages, create social media adaptations for X, LinkedIn, Facebook, Reddit, Threads…
geo-optimizer
Make content quotable and citable by AI answer engines (ChatGPT, Perplexity, Google AI Overviews) with answer-first structure, sourced stats, entities, and FAQ blocks. Use for GEO / AI-search optimization.
wp-article-writer
Write WordPress-ready SEO + GEO articles with search-intent structure, snippet targeting, and front matter that maps to the REST API. Use when drafting a blog post or article for WordPress.