Self-hosted X (Twitter) content management and AI-powered posting platform with MCP server support for programmatic access
SuperX defines 9 tools with complete JSON schemas and descriptions. Naming follows verb_noun convention (get_*, search_*, draft_*). Descriptions are present and contextual (avg ~120 chars), explaining WHAT and WHEN. However, parameter descriptions are sparse or missing for optional fields (e.g., 'days' in get_analytics lacks guidance on default behavior; 'status' enums lack explanation of what each state means). Output schemas are not documented, LLMs cannot predict response structure for chaining. Error handling is absent from tool definitions. One write tool (draft_post) lacks confirmation/dry-run pattern. Overall: solid foundation, but incomplete for production agent use.
Write a post in the user's voice about a topic. Returns the draft text and saves it to Ready to Post; does not publish it.
Follower, post, impression and engagement figures for a recent window.
The user's long-form Articles. These are separate from post drafts, the Ready to Post shelf, and the posting queue.
Mentions or topic-feed items waiting for a reply. Mentions are highest priority first; feed items are newest first.
The topic feeds the user follows, including whether each is active and whether it has ever synced.
Contacts found by Signals agents, optionally filtered by their relationship stage, best match first.
Output schemas not documented. LLMs cannot predict response structure, breaking tool chaining and forcing exploratory calls.
Optional parameter descriptions incomplete. 'days' in get_analytics states 'Defaults to 30' but does not explain what metric window it controls. 'status' enums lack explanation of state transitions.
draft_post (WRITE tool) lacks confirmation or dry-run pattern. Agents can draft posts without preview or approval step, risking unintended content creation.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 57 | 2026-07-28+ | v2 |
What is currently scheduled, drafted, published or failed.
Drafts waiting on the Ready to Post shelf. These are written but not yet approved, and are separate from the queue — a question about drafts, the shelf, or what is waiting for review is answered here, not by get_queue.
Search the library of high-performing posts. Use this to ground advice in what actually worked rather than guessing.
No error handling guidance in tool definitions. LLMs cannot recover from failures (e.g., 'search_inspiration' with no matching posts, 'get_leads' with invalid status).
Pagination not declared. Tools returning lists (get_queue, get_engagements, get_leads) lack limit/offset/cursor parameters, risking context window exhaustion on large result sets.