Eclipse Dispatch MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| ECLIPSE_DISPATCH_STATE | No | Path to the persistent state file (defaults to data/state.json beside the server) | data/state.json |
| ECLIPSE_DISPATCH_GHOST_URL | No | Base URL for the Ghost adapter endpoint; with none set, in-process fixture transports answer | |
| ECLIPSE_DISPATCH_BEEHIIV_URL | No | Base URL for the beehiiv adapter endpoint; with none set, in-process fixture transports answer |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| draft_createB | Create (or re-version) the canonical draft: title, body, subtitle, canonical_url (home), tags, audience, per-platform variants. Hash-bound: the returned content_hash is what approval binds to. Re-creating an existing draft_id makes a NEW version and resets approval to none. |
| draft_previewB | The review artifact: per-platform rendered previews + every adaptation diff against the canonical draft (variant overrides and platform constraints, e.g. Substack = paste pack, beehiiv = draft-only). |
| draft_approveA | Record owner approval of the current version. Returns an approval token bound to the content hash; any later edit invalidates it. |
| syndicate_draftsA | Fan the approved draft out to per-platform DRAFTS via the fixture adapters (beehiiv/Ghost draft APIs-shaped; Substack gets a paste pack — read-only adapter, never a write). Requires the approval token and an un-invalidated approval hash. There is no publish path. |
| analytics_reportB | D0 unified analytics (read-only): per-platform figures normalized by metric key, each carrying platform provenance + the platform's own definition. Combined figures appear ONLY where definitions are identical; differing definitions are flagged DEFINITIONS_DIFFER, never averaged. |
| receipt_listB | List hash-chained receipts (draft creations, approvals, syndications per platform, analytics runs) and verify the chain. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 6 tools
Each tool has a clearly distinct role in the draft lifecycle: creation, previewing, approval, syndication, analytics, and receipts. There is little risk of an agent confusing these operations because their boundaries are explicit, especially around approval and syndication requirements.
The set mixes naming conventions: draft_create/draft_preview/draft_approve use noun_verb with a shared draft_ prefix, syndicate_drafts uses verb_noun, and analytics_report/receipt_list use other noun-based patterns. All names are snake_case and readable, but the verb/noun order is not predictable across the set.
Six tools is well-scoped for a focused draft-to-syndication workflow with analytics and audit receipts. Each tool earns its place, and the set avoids unnecessary surface area.
The core lifecycle is covered: create/re-version, preview, approve, syndicate as platform drafts, report analytics, and verify receipts. Minor gaps remain, such as no explicit draft list/get or delete/archive operation, though re-creation handles versioning and publishing is intentionally excluded.