Seo Geo Qa
v2.1.0Check blog posts and articles before publishing. Finds broken links, weak sources, missing SEO elements, and citation problems. Use when: reviewing a draft,...
⭐ 0· 213·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (SEO content QA) match the included scripts and references. The scripts parse Markdown articles, verify links, perform SERP/competitor fetches, and generate reports — all consistent with an audit tool. The use of r.jina.ai as a fallback for fetching rendered pages is consistent with the stated need to bypass anti-bot protections for competitor pages.
Instruction Scope
Runtime instructions tell the agent to read the article file(s), run the included Python scripts, perform HTTP requests (DuckDuckGo, competitor pages, and optionally r.jina.ai), and write timestamped markdown/JSON reports under qa-reports/<article-slug> next to the article. Scripts look for workspace markers (AGENTS.md or .git) by walking up a few directory levels to locate roots; this is scope-relevant but only checks existence, not contents. Nothing in SKILL.md or the scripts instructs reading arbitrary unrelated user files or environment secrets.
Install Mechanism
There is no install spec (instruction-only skill with bundled scripts). No external binary downloads or package installs are declared. The only external tool dependency is curl (invoked for HTTP HEAD checks), which is reasonable for the stated checks.
Credentials
The skill declares no required environment variables, no credentials, and no config paths beyond an optional small JSON config. Network access is required (DuckDuckGo, target sites, and r.jina.ai) which is proportionate to a web-content QA tool. No broad or unrelated secrets are requested.
Persistence & Privilege
The skill does not request always:true and does not modify other skills or system-wide settings. It writes local report files (qa-reports/...) next to the article, which is expected for a QA runner. Autonomous invocation is allowed (platform default) and appropriate for a utility skill.
Assessment
This skill appears to do what it claims: it reads article Markdown, fetches pages (via urllib and curl), optionally uses r.jina.ai to render pages behind bot protection, and writes human-readable and JSON reports under qa-reports/<article-slug> next to your article. Consider the following before installing: 1) it requires network access and will fetch external URLs (including r.jina.ai) — if you need isolation, run it in a sandbox or disable Jina with --no-jina; 2) it writes report files adjacent to your content (qa-reports/...); ensure that write location is acceptable; 3) scripts check for .git or AGENTS.md up to a few parent folders to find workspace root — this only detects presence but you may want to run from the intended working directory; 4) review the source if you have strict privacy or corporate policy around external fetches or third-party services (r.jina.ai) before allowing the tool network access. Overall, there are no unexplained credential requests or attempts to access unrelated system secrets.Like a lobster shell, security has layers — review code before you run it.
latestvk97b4hxwd8hfh3zdy4z5da44g5844r8v
License
MIT-0
Free to use, modify, and redistribute. No attribution required.