oi.workflows.getReturn the reusable compiled prompt scaffold for a specific Oi workflow by workflowId. Prefer oi.workflows.use when the user has a concrete task.
| Name | Required | Description | Default |
|---|---|---|---|
| workflowId | Yes | The installed workflow identifier to load. |
| Name | Required | Description | Default |
|---|---|---|---|
| content | Yes | ||
| executionPlan | No |
Changes observed during successful MCP inspections. Dates show when Glama detected each change.
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint, idempotentHint, and destructiveHint, so the safe-read nature is fully covered. The description adds only that the result is a 'compiled prompt scaffold', which is light behavioral context. There is no contradiction with annotations, but no deeper disclosure of error or caching behavior is provided.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences with no filler. The primary action is front-loaded, and the sibling-routing guidance is placed second, making the description easy to parse at a glance.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a one-parameter, read-only, idempotent getter with an output schema and explicit sibling guidance, nothing essential is missing. The description correctly orients an agent to when to call this tool and what to expect from the result.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The schema has 100% coverage for the single workflowId parameter with a clear 'installed workflow identifier to load' description. The tool description repeats that it identifies by workflowId but adds no additional semantic detail beyond the schema. Baseline 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses a specific verb ('Return'), names the exact resource ('reusable compiled prompt scaffold for a specific Oi workflow'), and identifies the key discriminator (workflowId). It is clearly distinguished from siblings like oi.workflows.list ('specific' vs collection) and oi.workflows.use ('scaffold' vs running a task).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
It explicitly routes to oi.workflows.use when the user has a concrete task, giving a clear when-not condition. The phrase 'by workflowId' also implies the tool is for fetching one known workflow rather than searching or listing, covering the main alternative.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Resource prefixes (contexts, skills, guardrails, workflows, connections) make most tools clearly distinct, and parallel lifecycle verbs are scoped by resource name. The main ambiguity is within the guardrails publish/release/unpublish lifecycle and between brain.save-feedback, contexts.save-draft-feedback, and the two report tools, though detailed descriptions mostly resolve it.
Tools overwhelmingly follow an oi.<resource>.<verb> snake_case pattern, with create/get/list/update/use repeated consistently across resource types. Deviations include resource-less oi.recommend, noun-style oi.auth.whoami, and inconsistent release handling (separate guardrails.release vs action=release on contexts/skills update).
38 tools is too many for a single server, mainly because the same lifecycle pattern is repeated across Contexts, Skills, Guardrails, and Workflows. Each tool may be individually justifiable, but the set feels bloated and could benefit from consolidation or splitting into per-resource servers.
CRUD coverage is uneven: Guardrails have create/update/delete/list/get plus publish/unpublish/release, while Contexts, Skills, and Workflows lack any delete or archive tool, and Brain has only save-feedback with no read/update/delete path. Connections and auth are read/use-only, which may be intentional but leaves management actions absent.