agentleFS
try-eratry-era/geo-skills/skills/era/SKILL.md

era

Skill1 starsChanged 29 days agoMITView on GitHub

Entry point for Era's GEO and SEO skills. Use when the user asks what can be done for a brand or site in AI search, mentions GEO, AEO or AI visibility without naming a specific task, or wants a plan that chains several steps (audit, keywords, content calendar, review, publish). Checks the Era connection and routes to the right skill.

  • Reads credentials

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 3 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When the user already names a task ("audit this page", "approve next week's topics"), skip the router and run that skill directly.

    line 19
  2. Distinctive:

    Ask only for what is missing:

    line 24
  3. Very specific:

    Check the connection: Call whoami. If it fails or the era MCP server is missing, say so once, give the one-line fix, and continue with the …

    line 33
  4. Very specific:

    Resolve the brand: Call list_brands. Match the user's name against brand names and websites.

    line 44
  5. Distinctive:

    When routing, keep it to three lines: the brand/URL you resolved, the skill(s) you are running, and anything the user must decide first (a …

    line 81
  6. Very specific:

    Never start a skill that spends money or allowances (approve_topics, keyword volume checks, Suggest, test articles) from the router.

    line 88
  7. Very specific:

    Never guess a brand_id; never run a data skill against a brand you could not resolve.

    line 90
  8. Very specific:

    If whoami reports a read-only key and the chosen skill needs write access, say so before starting rather than failing halfway.

    line 91

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Era — router

Turn a loose request ("how do we show up in ChatGPT?", "do something for acme.com") into the right skill, or the right sequence of skills, and make sure Era is reachable before any data-backed step runs.

When to use

  • The user names a brand or a URL and a goal, but not a method.
  • The user asks what this plugin can do.
  • A job spans several skills (for example "audit the site, then plan content").

When the user already names a task ("audit this page", "approve next week's topics"), skip the router and run that skill directly.

Inputs

Ask only for what is missing:

  • Subject — a brand name (resolved through Era) and/or a URL (works without Era).
  • Goal — one of: see where we stand, beat a competitor, find what to write, fix a page, run the calendar, ship content to our site.
  • Window — days of data for reporting; default 30.

Steps

  1. Check the connection. Call whoami. If it fails or the era MCP server is missing, say so once, give the one-line fix, and continue with the skills that work on public pages only:

    claude mcp add --transport http era https://app.tryera.ai/api/mcp

    For automation without a browser sign-in, an API key works instead: --header "Authorization: Bearer $ERA_API_KEY". Never ask the user to paste a key into the chat.

  2. Resolve the brand. Call list_brands. Match the user's name against brand names and websites. One match: use its brand_id and say which brand you picked. Several: ask. None: the data skills are off the table for this brand; offer the public-page skills.
  3. Pick the route with the table below. Say in one sentence which skill(s) you will run and why, then run them. Chain at most four skills in one go; after that, stop and report.
  4. Hand off. Each skill produces its own deliverable. When chaining, pass forward only what the next skill needs (brand_id, the gap list, the chosen keyword cluster), not the whole previous report.

Routing table

The user wants…RunNeeds Era
to know how the brand is doing in AI answersai-visibility-report✓
to know why a competitor is named and they are notcompetitor-gap✓
to know which sites the engines trust in their categorycitation-sources✓
keywords or demand for AI searchai-keyword-research✓
a page that AI engines will quoteanswer-engine-contentoptional
to know how engines read a URL / a GEO or AEO auditgeo-auditoptional
an llms.txtllms-txt—
a brief for an articlecontent-brief✓
to see, rename, approve or pause planned articlescontent-calendar✓
to edit a generated article before it goes outarticle-review✓
articles delivered to a hand-built sitepublish-to-site✓
crawl, index, speed, canonical or sitemap problemstechnical-seo-audit—
JSON-LD / structured dataschema-markup—
title, meta, headings, internal links for one pageon-page-seo—

Common chains:

  • Where do we stand → ai-visibility-report → competitor-gap → citation-sources.
  • What should we write → competitor-gap → ai-keyword-research → content-brief.
  • Fix a page → geo-audit → on-page-seo → schema-markup → answer-engine-content.
  • Run the machine → content-calendar → article-review → publish-to-site.

Output

When routing, keep it to three lines: the brand/URL you resolved, the skill(s) you are running, and anything the user must decide first (a window, a spend). The chained skills then print their own reports in order. Finish with a short "next steps" list: the skills that would be useful next and why.

Guardrails

  • Never start a skill that spends money or allowances (approve_topics, keyword volume checks, Suggest, test articles) from the router. Route to the owning skill, which asks.
  • Never guess a brand_id; never run a data skill against a brand you could not resolve.
  • If whoami reports a read-only key and the chosen skill needs write access, say so before starting rather than failing halfway.

Era tools used

  • whoami (read) — connection and scope check.
  • list_brands (read) — brand resolution.

Without Era

Route to geo-audit, llms-txt, technical-seo-audit, schema-markup, on-page-seo and answer-engine-content. Mention once that the reporting and calendar skills need an Era workspace, and move on.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.