Borrowing it
Nothing to install: this file belongs to QexleOSS/seo-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/QexleOSS/seo-mcp/main/.agents/skills/seo-radar/SKILL.mdgit clone --depth 1 https://github.com/QexleOSS/seo-mcpWrote 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/qexleoss/seo-mcp/seo-radar)<a href="https://agentmods.dev/skills/qexleoss/seo-mcp/seo-radar"><img src="https://agentmods.dev/badge/skills/qexleoss/seo-mcp/seo-radar/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/qexleoss/seo-mcp/seo-radar"><img src="https://agentmods.dev/badge/skills/qexleoss/seo-mcp/seo-radar.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00046 | $0.00672 |
| Opus 5.5 | $0.00018 | $0.00269 |
| Sonnet 5.5 | $0.00009 | $0.00134 |
| Haiku 4.5 | $0.00005 | $0.00067 |
Grade A, and why
seo-radar scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SEO Radar
Turns three live readings into one file per niche per day:
- what people are searching for right now (
get_trending) - which keyword candidates the autocomplete engines accept (
get_keyword_research) - what the result page for a keyword looks like (
get_serp_snapshot)
Every number in the file names the source it came from. A trend is not a volume estimate, and the file keeps that difference visible.
Run it
bunx @qexle/seo-mcp radar --niche "<topic>" --out radar-$(date +%F).md
Without --out the markdown goes to stdout, which is what an agent session
wants. --seed "<keyword>" starts the keyword pass from a specific term
instead of the niche phrase, and --limit sets how many rows each section
keeps (default 8).
As an MCP server the same three tools are available directly:
{"tool": "get_trending", "arguments": {"niche": "seo tools", "geo": "US"}}
{"tool": "get_keyword_research", "arguments": {"seed": "technical seo audit", "depth": "quick"}}
{"tool": "get_serp_snapshot", "arguments": {"keyword": "technical seo audit"}}
Working method
- Pick the niche in the words a customer would type. "coffee shops in Berlin" beats "local coffee market".
- Read the trends section first. It says what is moving now, so the keyword pass asks about live demand rather than a frozen list.
- Use the keyword candidates as they are. The autocomplete rank is the engine's own order, so a candidate at rank 1 outranks a candidate at rank 6 without needing a score.
- Snapshot only the candidates that matter. Each snapshot is a live request.
- Write the file with the date in the name, so a later run can be compared against it.
Rules
- Name the source for every figure. The file already does, and a summary written from it must as well.
- Report the tools that came back silent. An engine that answered nothing is missing data, not zero demand.
- Keep the difficulty score labelled as an estimate. It is a score built from the visible result page, not a search volume.
- Do not add volume numbers. The keyless path has no volume source, and a made-up number is worse than a missing one.
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.
- 2d ago First seen · 68 lines · 46 tokens per session scan A 2eaa7de95470
seo-radar is a skill published in the GitHub repository QexleOSS/seo-mcp (0 stars, last pushed 3d ago), licensed MIT. It adds 46 tokens to every session and 672 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-11-04.
seo-radar compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| seo-radarQexleOSS/seo-mcp | Skill | 0 | 3d ago | MIT | A | 46 | Claude Code, Codex |
| gentech-blockrunBlockRunAI/blockrun-mcp | Skill | 389 | yesterday | MIT | A | 46 | Claude Code, Codex |
| portainer-mcp-hygieneportainer/portainer-mcp | Skill | 241 | 15d ago | MIT | A | 315 | Claude Code, Codex |
| hmem-releaseBumblebiber/hmem | Skill | 24 | 6d ago | MIT | A | 50 | Claude Code |
| disease-analysisdbcls/togomcp | Skill | 21 | 2d ago | MIT | A | 149 | Claude Code, Codex |
| research-article-analysisdbcls/togomcp | Skill | 21 | 2d ago | MIT | A | 147 | Claude Code, Codex |
| index-advisorDeepSQLAI/deepsql | Skill | 40 | 5d ago | Apache-2.0 | A | 32 | Claude Code, Codex |