Claude Cursor GitHub Copilot Agent

README

Imported from mykpono/ultimate-seo-geo/agents/README.md.

LLM Mart · 0 points · 9 0 0

What vetted this — trust report

Download mykpono-ultimate-seo-geo-agents_README.md-bb25619.zip · 0 KB

mykpono/ultimate-seo-geo 78 17 forks MIT Updated 3d ago

Part of mykpono/ultimate-seo-geo — 4 skills

Install

skills CLI npx skills add https://github.com/mykpono/ultimate-seo-geo/tree/main/agents/README.md

Git git clone https://github.com/mykpono/ultimate-seo-geo.git

The skills CLI installs just this skill, for any of its supported agents. Git is the plain clone.

README

Subagent capability definitions

Parallel workers for SEO audits are documented in PARALLEL-AUDIT.md (single file: scopes, scripts, references).

How platforms use this

Platform	Interpretation
Cursor	Host AI uses Task tool to spawn parallel subagents from `PARALLEL-AUDIT.md`
Claude Code	Can install excerpts under `~/.claude/agents/` with YAML frontmatter if desired
Copilot / Gemini / others	Host reads as context for how to parallelize script runs

Orchestration pattern

Run site_mapper.py to discover URLs when needed.
Spawn workers per row in PARALLEL-AUDIT.md (independent script groups only).
Merge results; deduplicate with finding_verifier.py.
Produce the unified report. The Health Score comes only from generate_report.py; without it, report "not scored" (§ 2).

See also references/procedures/21-script-toolbox.md (orchestrator-workers pattern, context compaction).

Files (ultimate-seo-geo)

README.md 1004 B

# Subagent capability definitions

Parallel workers for SEO audits are documented in **[`PARALLEL-AUDIT.md`](PARALLEL-AUDIT.md)** (single file: scopes, scripts, references).

## How platforms use this

| Platform | Interpretation |
|----------|----------------|
| **Cursor** | Host AI uses Task tool to spawn parallel subagents from `PARALLEL-AUDIT.md` |
| **Claude Code** | Can install excerpts under `~/.claude/agents/` with YAML frontmatter if desired |
| **Copilot / Gemini / others** | Host reads as context for how to parallelize script runs |

## Orchestration pattern

1. Run `site_mapper.py` to discover URLs when needed.
2. Spawn workers per row in `PARALLEL-AUDIT.md` (independent script groups only).
3. Merge results; deduplicate with `finding_verifier.py`.
4. Produce the unified report. The Health Score comes only from `generate_report.py`; without it, report "not scored" (§ 2).

See also `references/procedures/21-script-toolbox.md` (orchestrator-workers pattern, context compaction).