Draft, check and schedule social posts through KreatorMesh from your AI agent, plus a free skill on testing content hooks against your own audience.
- ✓Open-source license (MIT)
- ✓Actively maintained (<30d)
- ✓Clear description
- ✓Documented (README)
/plugin marketplace add orkuma-alex/kreatormesh-agent-mode
/plugin install kreatormesh-agent-modeResumen de Plugins
# KreatorMesh Agent Mode [](LICENSE) [](https://glama.ai/mcp/connectors/com.kreatormesh/kreatormesh) Give your AI agent the ability to draft, check and schedule social posts — and, more usefully, to find out what has actually worked on your audience before it writes anything. **Supports:** TikTok, Instagram, YouTube, X (Twitter), LinkedIn, Facebook, Threads, Pinterest, Bluesky, Google Business Profile. Built on [KreatorMesh](https://kreatormesh.com). ## Two skills, and one of them is free | Skill | Needs an account? | What it does | |---|---|---| | [`hook-testing-method`](skills/hook-testing-method/SKILL.md) | **No** | How to work out which opening lines hold attention on *your* audience, instead of following generic hook advice. Method only — a spreadsheet is enough. | | [`kreatormesh`](skills/kreatormesh/SKILL.md) | Yes | Draft, check against your learned rules, schedule, and review — through KreatorMesh. | The first is free and standalone on purpose. Most hook advice is an average taken across audiences that are not yours, and the fix is measurement rather than better advice. That skill explains how to do it by hand. It is genuinely useful without paying anyone. ## Install ### As skills (Claude Code, Cursor, Windsurf, Codex and others) ```bash npx skills add orkuma-alex/kreatormesh-agent-mode ``` Install just one: ```bash npx skills add orkuma-alex/kreatormesh-agent-mode --skill hook-testing-method ``` ### As an agent plugin The repo root is a portable [Agent Plugin](https://agent-plugins.org) (spec 1.0.0): `plugin.json` + `mcp.json` + `skills/`. Claude Code additionally reads `.claude-plugin/`. Both layouts sit side by side, so the same repo works either way. ### CLI only ```bash npx kreatormesh-cli setup --key km_live_... npx kreatormesh-cli accounts ``` Node 18+. No dependencies. ## Connecting You need a KreatorMesh account on the Creator plan or above and an API key from <https://kreatormesh.com/api-keys>. Keys are shown once. **MCP** — point your client at `https://api.kreatormesh.com/api/mcp` with the key as a bearer token, or set `KREATORMESH_API_KEY` and use the bundled `mcp.json`. **CLI** — `npx kreatormesh-cli setup --key km_live_...`, or set `KREATORMESH_API_KEY` in the environment (preferred in CI and agent sandboxes, since it leaves nothing on disk). ## Tools | Tool | Use it for | |---|---| | `list_connections` | account ids, handles and platforms. Call first. | | `list_hook_rules` | rules learned from this account's completed hook tests | | `check_draft` | check a caption against those rules before publishing | | `schedule_post` | save a draft, or schedule a post, with or without media (public https URLs) | | `list_posts` | drafts, scheduled, posted and failed. Filter with `status`, keep `limit` small. | | `get_post` | one post by id: caption, media, schedule, status, and per-platform links once published | | `cancel_post` | delete a post that has not gone out yet. Refuses published posts and hook-test posts. | | `get_performance` | metrics for published posts: totals across all of them, plus the newest `limit` in detail | | `list_evergreen` | posts that already performed and are being recycled | | `create_hook_test` | start a hook test. Schedules real posts, so confirm first. | The ordering is the point. `check_draft` between writing and scheduling is what makes this different from a cross-poster: it tells you which of your *measured* rules a draft breaks, and by how much. ## Known limits, stated up front - **Media is attached by URL.** `schedule_post` takes public https links to images or video, and the platform fetches them when the post publishes, so the link must still work then. Instagram, TikTok, YouTube and Pinterest need at least one; X, LinkedIn, Facebook, Threads, Bluesky and Google Business accept text alone. - **`cancel_post` only stops posts that have not gone out.** A published post is already live on the platform; deleting KreatorMesh's record would not take it down, so the tool refuses and says so. - **`check_draft` returns nothing until a hook test has completed** in the app. An empty result means "no rules learned yet", not "the draft is fine". The free `hook-testing-method` skill explains the method meanwhile. - **Hook tests need a platform that reports views:** X, YouTube, Instagram, Facebook, Threads, Pinterest and TikTok. Bluesky and LinkedIn report likes and comments only, so tests are refused there. - **API keys cannot manage API keys.** Listing, creating and revoking need a signed-in session, so a leaked key cannot mint more or revoke the one you are using to investigate it. - **A key acts as its owner.** Anything these tools can do, that person could do. ## Security Keys are stored hashed server-side — the plaintext is shown once and cannot be retrieved, so a lost key is replaced rather than recovered. Revoke at <https://kreatormesh.com/api-keys>; revocation takes effect on the next request. The CLI saves a key to `~/.kreatormesh/config.json` with owner-only permissions where the platform supports it. `KREATORMESH_API_KEY` takes precedence and writes nothing to disk. ## Registry listing `server.json` is the official MCP Registry entry; see [PUBLISHING.md](PUBLISHING.md). ## Licence MIT. See [LICENSE](LICENSE).
Lo que la gente pregunta sobre kreatormesh-agent-mode
¿Qué es orkuma-alex/kreatormesh-agent-mode?
+
orkuma-alex/kreatormesh-agent-mode es plugins para el ecosistema de Claude AI. Draft, check and schedule social posts through KreatorMesh from your AI agent, plus a free skill on testing content hooks against your own audience. Tiene 0 estrellas en GitHub y su última actualización registrada es del 2026-10-01.
¿Cómo se instala kreatormesh-agent-mode?
+
Puedes instalar kreatormesh-agent-mode clonando el repositorio (https://github.com/orkuma-alex/kreatormesh-agent-mode) o siguiendo las instrucciones del README en GitHub. ClaudeWave también te ofrece bloques de instalación rápida en esta misma página.
¿Es seguro usar orkuma-alex/kreatormesh-agent-mode?
+
Nuestro agente de seguridad ha analizado orkuma-alex/kreatormesh-agent-mode y le ha asignado un Trust Score de 87/100 (tier: Trusted). Revisa el desglose completo de comprobaciones superadas y flags en esta página.
¿Quién mantiene orkuma-alex/kreatormesh-agent-mode?
+
orkuma-alex/kreatormesh-agent-mode es mantenido por orkuma-alex. La última actividad registrada en GitHub es del 2026-10-01, con 0 issues abiertos.
¿Hay alternativas a kreatormesh-agent-mode?
+
Sí. En ClaudeWave puedes explorar plugins similares en /categories/plugins, ordenados por popularidad o actividad reciente.
Despliega kreatormesh-agent-mode en tu cloud
Lleva este repo a producción en minutos. Cada plataforma genera su propio entorno con variables de entorno editables.
¿Mantienes este repo? Añade un badge a tu README
Pega el badge en tu README de GitHub para mostrar que está auditado por ClaudeWave. Cada badge enlaza de vuelta a esta página y muestra el Trust Score actual.
[](https://claudewave.com/repo/orkuma-alex-kreatormesh-agent-mode)<a href="https://claudewave.com/repo/orkuma-alex-kreatormesh-agent-mode"><img src="https://claudewave.com/api/badge/orkuma-alex-kreatormesh-agent-mode" alt="Featured on ClaudeWave: orkuma-alex/kreatormesh-agent-mode" width="320" height="64" /></a>Más Plugins
Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.
AI agent skill that researches any topic across Reddit, X, YouTube, HN, Polymarket, and the web - then synthesizes a grounded summary
Write HTML. Render video. Built for agents.
Agent skill that removes signs of AI-generated writing from text
Academic Research Skills for Claude Code: research → write → review → revise → finalize
Create beautiful slides on the web using a coding agent's frontend skills