seo-service
97humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
An SEO service that checks blog pages against rules built from RSS articles about search optimisation. It can connect to WordPress to read and update posts.
Drewgent configuration for opencode — agent profiles, skills, scripts, and tooling
This repository also configures its own agents. See what opencode-drewgent tells them →
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
An SEO service that checks blog pages against rules built from RSS articles about search optimisation. It can connect to WordPress to read and update posts.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
A guide to traditional SEO and GEO, which means preparing content to be found, understood, and cited by AI search systems. It applies these checks to WordPress sites, Cloudflare Workers, and an Obsidian vault.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
A website-audit guide that checks a URL against The Website Specification, a collection of website requirements and recommendations. It covers areas such as accessibility, SEO, security, and readiness for AI agents.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
A guide for reviewing which skills in an agent's skills folder are active and which appear unused. The available description does not provide enough detail about its complete audit process.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
Interact with X/Twitter via the x-cli terminal client using official X API credentials. Use for posting, reading timelines, searching tweets, liking, retweeting, bookmarks, mentions, and user lookups.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
X/Twitter via xurl CLI: post, search, DM, media, v2 API.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
Pattern for organizing CF Workers into router → controllers → engine → utils layers. When a worker exceeds 300 lines, split into thin router + single-responsibility controllers + pure computation engine + shared utilities.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex needs its repo
Use when refactoring a Cloudflare Workers monolith into controller-based architecture. Covers dead code detection, master data consolidation, cross-source sync, and live dev-server testing.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex needs its repo
Class-level skill for integrating separate codebases (analysis engines, NAS controllers, shared utilities, LLM callers) into a single coherent Cloudflare Workers project.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
Debug and operate Cloudflare Workers deployments — secrets management, local-vs-deployed diagnostics, outbound API troubleshooting, and deployment workflow.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
Find and merge duplicate data structures, constants, and logic across architectural boundaries — frontend↔backend, separate workers, independent modules. Create single sources of truth.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
Safe, incremental codebase refactoring methodology — dead code removal, data consolidation, CSS chain unification, and PII protection. One change at a time, verified with a running dev server.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
Map dependency graphs, detect dead code, and execute one-change-at-a-time refactoring with live dev server verification.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
Reduce LLM token spend on cron + background + kanban-worker without touching interactive (CLI/ACP/Discord) paths. Inventory → classify → apply config/scheduler/worker patches → verify with hard evidence.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
Evaluate a third-party GitHub repo or external tool for {{AGENTNAME}} integration. Use raw markdown curl + GitHub API to bypass UI, identify algorithm taxonomy, map to {{AGENTNAME}} hot spots, score 3-4 integration modes, recommend POC-first.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
Agent-level multi-model deliberation on OpenCode Go. Uses task() to simulate OpenRouter Fusion's panel+judge pipeline — parallel subagents with distinct perspectives, then structured synthesis. Zero marginal cost (OpenCode Go subscription). Reach for it when a single model isn't enough — research, expert critique…
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex needs its repo
Extract modules from gateway/run.py (9,876 lines) into isolated files under gateway/. Covers stdlib name collision, runner circular reference, mock fixture sync order, and honest multi-session QA verdicts.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
Session-specific details from the 2026-06-15 NAS Huly self-host session — the bugs that weren't bugs, the credential masking traps, and the actual working path.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex needs its repo
Safe, incremental codebase refactoring with dependency graph analysis and live dev server verification.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
A guide to adding Korean online payments to a plain JavaScript web app through PortOne, a service that connects several Korean payment providers through one SDK.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
A step-by-step process for changing the default language model or provider across code, configuration, documentation, and catalog entries.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
Cost-aware multi-tier model selection strategy. Assigns models to roles (main agent, subagent, auxiliary) by capability tier under subscription billing. Defines the 3-tier flash/pro/max pyramid and the escalation path. Supersedes the per-call era cost-optimization-background-llm skill.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
When a multi-page app (MPA) loses UX continuity (modals/sidebars/tabs reset on navigation), bridge local state to URL query params so each state is shareable, bookmarkable, and back-button-restorable — without changing visual design or feature behavior.
humanerd-drew/opencode-drewgent
Skill Claude CodeCodex
Use this skill when the user wants a plan instead of execution.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: