Tulina

oto_function

FUNCTIONS: pure code stored and run by Oto for an exact, deterministic calculation (JSON in → JSON result, warnings and files out). A function reads nothing from Oto, calls no network and sees no secret: pass it everything it needs in input. op=run (slug, input, project = where to store the files it produces → result, warnings, files with a signed download_url) runs the PUBLISHED version only · op=list · op=get (slug, optional version: default the published one, else the latest) · op=versions (history, without code) · op=create (slug, title, description, sources, entrypoint, requirements, note → version 1, PROPOSED) · op=propose (slug, expected_version = the latest version you read, sources, entrypoint, requirements, note → next version, PROPOSED). sources is {filename: content} — flat .py modules (tests are test_*.py, plain functions named test_*) and .json data. A proposed version never runs until it is tested and published (op=test/publish/refuse, by a person in scope: never from a hosted agent). Scope: your active org (default) or scope='user'. PROVISIONAL surface.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "op": "<op>"
    },
    "name": "oto_function"
  }
}

Input Schema

Name	Required	Description
`op`	Yes
`org`	No
`_org`	No
`note`	No
`slug`	No
`input`	No
`scope`	No
`title`	No
`project`	No
`sources`	No	Files of this version, {filename: content}: flat `.py` modules (tests are `test_*.py`) and `.json` data files.
`version`	No
`entrypoint`	No	`module:function`, called with the input and returning {result, warnings, files}.
`description`	No
`requirements`	No
`expected_version`	No

Schema Changelog

Changes observed during successful MCP inspections.

No schema history has been recorded yet.

TDQS

A4.5/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations supplied, the description carries the full behavioral burden and does it well: discloses the sandbox guarantees (reads nothing from Oto, no network, no secrets, everything must arrive via `input`), that op=run only executes the PUBLISHED version, the proposed→tested→published lifecycle, and that publish/refuse must be performed by a person, never a hosted agent.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loads the concept in the first sentence and then enumerates ops with a consistent '·' delimiter, so every clause carries information. It is extremely dense — a single packed paragraph with heavy parenthetical asides — which costs some scannability but not correctness.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Covers the mutation and run semantics thoroughly for a 15-parameter multiplexer, including the shape of run's return values (result, warnings, files with signed download_url). However, return shapes for the other ops (get, versions, list) are not described and there is no output schema, leaving the read-side behavior partly unspecified.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is only 13%, so the description must compensate and largely does: it defines the meaning of slug, input, project (where produced files land), version (defaults to published, else latest), sources (flat .py plus test_*.py and .json), entrypoint, requirements, note, expected_version and scope (org default vs user). Only minor gaps remain, such as org/_org and title not being explained.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('pure code stored and run by Oto for an exact, deterministic calculation') plus the contract (JSON in → JSON result, warnings and files out). This clearly separates it from siblings like oto_shell or oto_recipe without requiring the schema to be opened.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Effectively routes the agent across all nine ops with terse definitions (run/list/get/versions/create/propose/test/publish/refuse) and states the hard gate that a proposed version never runs until tested and published by a person in scope. It does not explicitly compare against alternative siblings (e.g. oto_shell or oto_recipe) for similar calculation tasks, so the exclusion guidance is implicit rather than stated.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Tulina tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?