# Rhylthyme (remote · mcp.rhylthyme.com) Real-time multi-track schedules for cooking, lab protocols, events and workouts. Validate and share. - Trust score: 78/100 (medium) - Change this week: +4 - Registry status: active - Liveness: live - Owner verified: no - Last scored: 2026-09-24 ## Components - remote · `mcp.rhylthyme.com`: 78/100 (this document), [markdown](https://verifymcp.io/servers/com-rhylthyme-rhylthyme/mcp.md), [page](https://verifymcp.io/servers/com-rhylthyme-rhylthyme/mcp) ## Channel facts - Endpoint: `https://mcp.rhylthyme.com/mcp` - Transports: `streamable-http` - Auth: `none` - Version: `1.3.0` ## Trust breakdown How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog) Scored 2026-09-24. - **Endpoint Security**: 80/100 - The endpoint's TLS certificate is valid, in date, and uses a strong key. - No authorisation is required to call this server. Every tool declares its destructiveHint and none is destructive, so open access doesn't expose one. - HTTPS is enforced; there's no plaintext access path. - The HSTS (Strict-Transport-Security) header is present. - DNSSEC check failed: this domain isn't protected by DNSSEC. - **Transport & Reachability**: 100/100 - Verified streamable-http transport via a live MCP handshake. - **Schema Quality & AI Usability**: 86/100 - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name). - AI-judged instruction clarity (excellent). - Context-footprint check failed: tool/resource definitions use about 3397 tokens (~117/item across 29 items; 19 tools + 10 resources), over budget; trim descriptions and params. - Usage-examples check failed: none of the tools include examples. - **Stability & Change Management**: 23/100 - Stability observed for 7 of 30 days with no destabilising changes; credit accrues until the full window elapses. - **Tool Coverage**: 96/100 - 100% of tools have a non-trivial description (not blank, and not just the tool's name). - 88% of tool parameters carry a description. - Structured output schemas are declared (21% of tools); any adoption earns full credit. - **Tool Safety**: 100/100 - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured. - All 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation. - An AI judge read all 21 captured unit(s) of tool text and found none that tries to manipulate the model reading it. - **Capabilities**: 100/100 - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28. ## Install ### How do I install the Rhylthyme MCP server? Rhylthyme is a hosted endpoint at https://mcp.rhylthyme.com/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation. ### Claude ```bash claude mcp add --transport http com-rhylthyme-rhylthyme 'https://mcp.rhylthyme.com/mcp' ``` ### Cursor ```json { "mcpServers": { "com-rhylthyme-rhylthyme": { "url": "https://mcp.rhylthyme.com/mcp" } } } ``` ### VS Code ```json { "servers": { "com-rhylthyme-rhylthyme": { "type": "http", "url": "https://mcp.rhylthyme.com/mcp" } } } ``` ### Codex ```toml [mcp_servers.com-rhylthyme-rhylthyme] url = "https://mcp.rhylthyme.com/mcp" ``` ### opencode ```json { "$schema": "https://opencode.ai/config.json", "mcp": { "com-rhylthyme-rhylthyme": { "type": "remote", "url": "https://mcp.rhylthyme.com/mcp", "enabled": true } } } ``` ### OpenClaw ```bash openclaw mcp add com-rhylthyme-rhylthyme --url 'https://mcp.rhylthyme.com/mcp' --transport streamable-http ``` ### Hermes ```yaml mcp_servers: com-rhylthyme-rhylthyme: url: "https://mcp.rhylthyme.com/mcp" ``` ### Netclaw ```json { "McpServers": { "com-rhylthyme-rhylthyme": { "Transport": "http", "Url": "https://mcp.rhylthyme.com/mcp" } } } ``` ### Vellum ```bash assistant mcp add com-rhylthyme-rhylthyme -t streamable-http -u 'https://mcp.rhylthyme.com/mcp' ``` ### Other ```json { "mcpServers": { "com-rhylthyme-rhylthyme": { "type": "http", "url": "https://mcp.rhylthyme.com/mcp" } } } ``` The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs. ## Changelog Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims. ### 2026-09-24 (score 78, 0) - [functional] New tool “review_program” - [cosmetic] “analyze_schedule” reworded the description of “history” - [cosmetic] “analyze_schedule” reworded the description of “predictionContext” ### 2026-09-22 (score 78, +1) No change was recorded against any check on this day. Stability & Change Management went from 13 to 17. That category is still filling its 30-day observation window: 4 days of observed history at the previous scan, 5 at this one. The score rises as the window fills, whether or not the server changes. ### 2026-09-21 (score 77, +1) - [security] The server rewrote its instructions, which are the text every model session reads - [security] Tool “analyze_schedule” rewrote its description, which is the text the model reads - [security] Tool “calibrate_program” rewrote its description, which is the text the model reads - [security] Tool “load_program” rewrote its description, which is the text the model reads - [security] Tool “create_environment” rewrote its description, which is the text the model reads - [security] Tool “get_renderer_source” rewrote its description, which is the text the model reads - [security] Tool “import_from_source” rewrote its description, which is the text the model reads - [security] Tool “import_text” rewrote its description, which is the text the model reads - [security] Tool “list_my_programs” rewrote its description, which is the text the model reads - [security] Tool “list_public_runs” rewrote its description, which is the text the model reads - [security] Tool “list_runs” rewrote its description, which is the text the model reads - [security] Tool “load_public_recipe” rewrote its description, which is the text the model reads - [security] Tool “load_run” rewrote its description, which is the text the model reads - [security] Tool “login” rewrote its description, which is the text the model reads - [security] Tool “preview_timeline” rewrote its description, which is the text the model reads - [security] Tool “save_program” rewrote its description, which is the text the model reads - [security] Tool “search_public_recipes” rewrote its description, which is the text the model reads - [security] Tool “validate_program” rewrote its description, which is the text the model reads - [security] Tool “visualize_schedule” rewrote its description, which is the text the model reads - [functional improvement] Schema quality: 188 → 116 - [functional] Destructive annotations: pass → 100 - [functional] New resource “tool-guide” - [cosmetic] “analyze_schedule” reworded the description of “finishAt” - [cosmetic] “analyze_schedule” reworded the description of “history” - [cosmetic] “analyze_schedule” reworded the description of “predictionContext” - [cosmetic] “analyze_schedule” reworded the description of “program” - [cosmetic] “analyze_schedule” reworded the description of “program_id” - [cosmetic] “analyze_schedule” reworded the description of “startAt” - [cosmetic] “analyze_schedule” reworded the description of “token” - [cosmetic] “analyze_schedule” reworded the description of “useDurations” - [cosmetic] “calibrate_program” reworded the description of “accept” - [cosmetic] “calibrate_program” reworded the description of “history” - [cosmetic] “calibrate_program” reworded the description of “k” - [cosmetic] “calibrate_program” reworded the description of “program” - [cosmetic] “calibrate_program” reworded the description of “program_id” - [cosmetic] “calibrate_program” reworded the description of “since” - [cosmetic] “calibrate_program” reworded the description of “token” - [cosmetic] “import_from_source” reworded the description of “enrich” - [cosmetic] “import_from_source” reworded the description of “query” - [cosmetic] “import_from_source” reworded the description of “text” - [cosmetic] “import_from_source” reworded the description of “token” - [cosmetic] “import_text” reworded the description of “deadline” - [cosmetic] “import_text” reworded the description of “environmentType” - [cosmetic] “import_text” reworded the description of “hints” - [cosmetic] “import_text” reworded the description of “text” - [cosmetic] “import_text” reworded the description of “token” - [cosmetic] “list_my_programs” reworded the description of “token” - [cosmetic] “list_public_runs” reworded the description of “limit” - [cosmetic] “list_public_runs” reworded the description of “program” - [cosmetic] “list_public_runs” reworded the description of “program_hash” - [cosmetic] “list_runs” reworded the description of “program_id” - [cosmetic] “list_runs” reworded the description of “token” - [cosmetic] “load_program” reworded the description of “program_id” - [cosmetic] “load_program” reworded the description of “token” - [cosmetic] “load_run” reworded the description of “run_id” - [cosmetic] “load_run” reworded the description of “token” - [cosmetic] “login” reworded the description of “token” - [cosmetic] “preview_timeline” reworded the description of “program” - [cosmetic] “preview_timeline” reworded the description of “run” - [cosmetic] “save_program” reworded the description of “program” - [cosmetic] “save_program” reworded the description of “token” - [cosmetic] “search_public_recipes” reworded the description of “environment” - [cosmetic] “search_public_recipes” reworded the description of “query” - [cosmetic] “validate_program” reworded the description of “program” - [cosmetic] “visualize_schedule” reworded the description of “allowInvalid” - [cosmetic] “visualize_schedule” reworded the description of “program” - [cosmetic] “list_my_programs” made “token” optional - [cosmetic] “load_program” made “token” optional - [cosmetic] “save_program” made “token” optional ### 2026-09-20 (score 76, +1) No change was recorded against any check on this day. Stability & Change Management went from 7 to 10. That category is still filling its 30-day observation window: 2 days of observed history at the previous scan, 3 at this one. The score rises as the window fills, whether or not the server changes. ### 2026-09-18 (score 75, +1) - [functional improvement] Stability: unverified → 0.03 ### 2026-09-17 (score 74) First indexed and scored. ## MCP tools (19) ### `validate_program` (~293 tokens) Validate a program Check a program before publishing or saving it: ids, dangling or cyclic triggers, steps overlapping in a track, tasks with no resourceConstraint, durations, replicates/instances misuse. Every finding has a code, a message and a fix; apply the fixes and re-run until valid. Pure computation. Input parameters: - `program` (object, required): Program JSON: {programId, name, environmentType?, actors?, tracks: [{trackId, name, steps: [{stepId (unique program-wide), name, task?, duration, startTrigger, replicates?, choice?}]}], resourceConst… Output parameters: - `errors` (array) - `info` (array) - `stats` - `valid` (boolean) - `warnings` (array) ### `analyze_schedule` (~307 tokens) Analyze schedule timing Put a program on the clock: each step's start and end, total length, the critical path and what gates it, resource conflicts, slack per track. Pass finishAt to answer 'when do I start X so everything is ready at 6pm?'. Pure computation. With `history` (or `program_id`) it also predicts durations from recorded runs; see rhylthyme://guide/tools. Input parameters: - `finishAt` (string): ISO 8601 datetime everything must END at, e.g. '2026-11-26T18:00:00-05:00'. Start times are worked backwards from it. - `history` (array): Recorded runs of this program (from load_run); adds predicted durations. - `predictionContext` (object): Optional: environmentId, userTags (variance factors), userId, programVersion, minIdentical, minModel, corrThreshold, verdicts. See rhylthyme://guide/tools. - `program` (object, required): Program JSON; shape under validate_program. - `program_id` (string): Saved program UUID: loads the user's own recorded runs as the history. - `startAt` (string): ISO 8601 start datetime. Ignored with finishAt. - `token` (string): Access token from the login tool. Omit when the account is connected. - `useDurations` (string): planned (default): analyse the program as authored. predicted: as the history says it will run. Output parameters: - `actorPeak` (object) - `bindingConstraints` (array) - `criticalPath` (array) - `durationsUsed` (string) - `inFlight` (array) - `makespan` (string) - `makespanSeconds` (number) - `name` - `plannedMakespan` (string) - `plannedMakespanSeconds` (number) - `predictedCriticalPath` (array) - `predictedMakespan` (string) - `predictedMakespanSeconds` (number) - `programId` - `resourceConflicts` (array) - `steps` (array) - `tracks` (array) - `validation` (object) - `wallClock` ### `visualize_schedule` (~117 tokens) Publish a live timeline Publish a program as a live timeline and return its URL. This is the deliverable for ANY schedule you produce ('plan this so it all finishes at 6', 'when do I start each thing'): never describe a schedule in prose. Validates first and refuses an invalid program with fix hints. Also returns an ASCII Gantt and an itinerary to quote. Input parameters: - `allowInvalid` (boolean): Publish despite validation errors. Only when the user explicitly wants an imperfect draft shared. - `program` (object, required): Program JSON; shape under validate_program. Output parameters: - `imageUrl` - `makespanSeconds` (number) - `shareId` - `url` (string) - `warnings` (array) ### `import_from_source` (~202 tokens) Import from an external source Import a recipe or protocol from a URL or service into a program. source: spoonacular, themealdb (recipes); protocolsio, opentrons, benchling (lab); cooklang (.cook URL). action: search (no login), import (query = URL or id), random. import, random and benchling need the account. `enrich: true` splits an import into parallel tracks. Then call visualize_schedule. Input parameters: - `action` (string, required) - `enrich` (boolean): With action='import': split the import into parallel tracks with cross-track triggers. One model call, capped per day. - `query` (string): Keywords (search), or the URL or id (import). - `source` (string, required) - `text` (string): Pasted source (an Opentrons .py) instead of a URL. - `token` (string): Access token from the login tool. Omit when the account is connected. ### `import_text` (~210 tokens) Import pasted text Turn pasted text (a recipe, protocol, run sheet or training plan) into a validated multi-track program on the server, in four model turns, with a table of the source span each step came from. Use when the user pastes the steps and no URL importer fits. Needs the user's Rhylthyme account. Capped per day; you can instead author the program yourself and call validate_program. Input parameters: - `deadline` (string): When everything must be finished, if the user said: '18:00', 'dinner at six', an ISO datetime. - `environmentType` (string, required): kitchen, lab, events, gym or generic. - `hints` (string): Equipment and people limits in the user's words, e.g. 'one oven, two burners, one cook'. - `text` (string, required): The source text, all of it, as the user wrote or pasted it. - `token` (string): Access token from the login tool. Omit when the account is connected. ### `create_environment` (~93 tokens) Define equipment limits Describe a workspace's equipment limits (one oven, two centrifuges, one stage) as resourceConstraints to copy into a program, so steps that share equipment are scheduled around each other. Input parameters: - `actors` (integer): How many people can work at once (default 1). - `description` (string) - `name` (string, required) - `resourceConstraints` (array, required) - `type` (string, required) ### `login` (~82 tokens) Connect Rhylthyme account For apps that cannot connect the Rhylthyme account directly: returns a sign-in URL, then accepts the pasted token. Only the user's own saved programs, runs and imports need it; everything public works without. Tokens last about an hour. Input parameters: - `token` (string): The token from the login page. Leave empty on the first call to get the sign-in URL. ### `list_my_programs` (~46 tokens) List my saved programs List the user's saved programs. Needs the user's Rhylthyme account. Input parameters: - `token` (string): Access token from the login tool. Omit when the account is connected. ### `load_program` (~70 tokens) Open a saved program Open one of the user's saved programs by id: summary plus live-timeline URL. Needs the user's Rhylthyme account. Input parameters: - `program_id` (string, required): Program UUID from list_my_programs. - `token` (string): Access token from the login tool. Omit when the account is connected. ### `list_runs` (~83 tokens) List recorded runs List recorded runs of a saved program: when, how it ended, actual against planned length. For 'how long did this take last time?'. Needs the user's Rhylthyme account. Input parameters: - `program_id` (string, required): Program UUID from list_my_programs. - `token` (string): Access token from the login tool. Omit when the account is connected. ### `review_program` (~147 tokens) Review an imported program A model reads an imported program against its source and reports what looks wrong: a default duration on "simmer for an hour", a step the importer dropped, steps chained that could overlap, a total that disagrees with the source. Changes nothing; returns findings to show the user. Needs the user's Rhylthyme account. Input parameters: - `program` (object, required): Program JSON; shape under validate_program. - `source_text` (string): The source text (recipe, protocol), if you have it. - `source_url` (string): Its URL, when the text is not to hand. - `token` (string): Access token from the login tool. Omit when the account is connected. ### `load_run` (~77 tokens) Open a recorded run Open one recorded run: planned against actual timing per step, what ended each step, pauses. Shows where a plan drifts. Needs the user's Rhylthyme account. Input parameters: - `run_id` (string, required): Run UUID from list_runs. - `token` (string): Access token from the login tool. Omit when the account is connected. ### `calibrate_program` (~218 tokens) Propose durations from recorded runs Propose durations for a saved program from its recorded runs: median as the default, P10 to P90 as the range, never narrower than the author's, with the evidence per step and the effect on total length. Saves nothing: pass `accept` to get the calibrated program, then save_program. Needs the user's Rhylthyme account. Input parameters: - `accept`: Step ids to accept, or "all": also returns the calibrated program. Nothing is saved either way. - `history` (array): Run records to use instead of the stored ones. - `k` (integer): Measurements a step needs for a proposal (default 5). - `program` (object): Program JSON to calibrate; omit to use program_id's saved JSON. - `program_id` (string): Saved program UUID whose recorded runs are the evidence. - `since` (string): Only runs started on or after this date (YYYY-MM-DD or ISO). - `token` (string): Access token from the login tool. Omit when the account is connected. ### `list_public_runs` (~112 tokens) List contributed runs List anonymous runs other people contributed for one exact program version: actual against planned length, and their conditions. For 'how long does this really take?'. No login needed. Pass `program` or `program_hash`. Input parameters: - `limit` (integer): Runs to return, newest first (default 50). - `program` (object): Program JSON to look up instead; hashed here. - `program_hash` (string): Program hash, 'sha256:' + 64 hex (a run record's programVersion). ### `search_public_recipes` (~94 tokens) Search the public catalog Search the public catalog of recipes, lab protocols, event templates and workouts by name or keyword. Returns ids, names, descriptions and URLs; no login. Pass an id to load_public_recipe. Input parameters: - `environment` (string): Which catalog: kitchen (default), laboratory, event or gym. - `limit` (integer) - `query` (string): Keyword; matches name and description. Empty returns the newest entries. Output parameters: - `count` (number) - `environment` - `query` (string) - `results` (array) ### `load_public_recipe` (~46 tokens) Open a public program Open a public catalog entry by id: summary plus live-timeline URL (no visualize_schedule call needed). Input parameters: - `program_id` (string, required): Program UUID (from search_public_recipes) ### `save_program` (~72 tokens) Save to my account Save a program to the user's account ('save this for later'); the same programId updates it. Needs the user's Rhylthyme account. Input parameters: - `program` (object, required): Program JSON; shape under validate_program. - `token` (string): Access token from the login tool. Omit when the account is connected. ### `preview_timeline` (~111 tokens) Preview timeline image Return a static Gantt image of a program, for 'show me a preview' or 'a picture of the timeline'. Image and caption only; use visualize_schedule for the live URL. Planned versus actual: pass `run` (from load_run) to draw the real bars over the plan. Input parameters: - `program` (object, required): Program JSON; shape under validate_program. - `run` (object): A recorded run of the same program (the `run` object from load_run): draws planned versus actual. ### `get_renderer_source` (~61 tokens) Get timeline renderer source Return the open-source timeline renderer's JavaScript (Apache-2.0, ~9KB, no dependencies) to paste into a