SomedayX
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| SOMEDAYX_HOME | No | Keep data somewhere else (e.g. for testing with an empty library). By default, saves, images, models and search index live in ~/Library/Application Support/SomedayX, which is where the MCP server reads from. Optional; normally not needed. |
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 | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_savesA | Search the user's saved X posts (bookmarks and likes) by meaning and keywords, including text inside
images and full X Articles. Describe what they remember in plain words ("that thread about pricing").
X-style operators also work inside the query: from:handle, is:bookmark, is:like, has:image, has:video,
has:link, has:article, year:2025, since:2025-03, until:2025-06. Dates are when the post was published.
Returns up to |
| recent_savesA | The user's most recently saved X posts, newest first, optionally filtered (source, author, has:image/video/link/article, publish dates since/until as YYYY, YYYY-MM or YYYY-MM-DD). Max 50. |
| get_postA | One saved post in full, by its id or x.com/twitter.com URL: complete text, text read from its images, the post it quotes, the full text of an X Article, and the links it contains. |
| library_overviewA | A summary of the user's saved X library: how many bookmarks and likes, the date range, the people they save most, and their most common topics. Useful before answering broad questions. |
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 4 tools
Each tool targets a distinct mode: semantic search, chronological listing, single-post retrieval, and aggregate stats. search_saves and recent_saves both return filtered lists of saved posts, but their descriptions clearly separate semantic ranking from newest-first ordering.
All names use consistent snake_case and readable wording, but the verb_noun pattern is not universal: get_post and search_saves lead with verbs, while recent_saves and library_overview are noun/qualifier-based. The deviation is minor and still predictable.
Four tools is a tight, well-scoped surface for a read-oriented saved-posts library, with each tool covering a distinct access pattern (search, recency, detail, overview). Nothing feels redundant or missing at the count level.
The read surface is solid: search with operators, recency listing, full-post detail, and library stats cover the main retrieval workflows. Gaps are pagination beyond limit and any write operations (save/unsave/tag), which agents can largely work around.