Greenhouse Ops Control Plane

by app.vercel.ats-ops-control-plane-gateway

create_standing_query

Save a buyer-defined standing query (custom recipe): a named recruiting question plus the prompt and read-tool list the model should use to answer it. Validates the definition against this connector's registered read tools and rejects any write tool. Hosted deployments persist it per license and surface it in get_control_plane_capabilities under custom_recipes; the connector itself never executes recipes.

Input Schema

TableJSON Schema

Name	Required	Description
`name`	Yes	Short human-readable title (3-120 chars).
`reason`	Yes	Human-readable audit reason for this change. Must be at least 12 characters.
`dry_run`	No	When true, validate and preview the change without sending a write request.
`enabled`	No	Disabled queries stay saved but are not surfaced.
`summary`	Yes	What the query computes and what the output looks like (20-2000 chars).
`actor_id`	Yes	Greenhouse user ID to send with this write.
`recipe_id`	Yes	Lowercase kebab-case slug identifying this standing query (3-64 chars). Re-using an existing slug replaces that saved query.
`required_tools`	Yes	Read tools (list_/get_) this query uses. Write tools are rejected.
`prompt_markdown`	Yes	The full recipe spec the model executes: tool choreography, computation, output format, traps (40-32000 chars). Stored and returned verbatim.
`example_question`	Yes	The natural-language question a buyer would ask to run this (10-500 chars).

TDQS

A4.3/5.0

Behavior5/5

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

With no annotations provided, the description carries the full burden. It covers validation (rejects write tools, validates against registered read tools), persistence (per license, surfaced via get_control_plane_capabilities), and execution model (connector never executes). Also notes slug reuse replaces the saved query and disabled queries are saved but hidden.

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?

The description is a single paragraph of four sentences, front-loading the primary action. Every sentence adds value, though it could be slightly more concise by combining the last two sentences.

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

Completeness4/5

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

The description covers validation, persistence, and execution model, and explains the purpose well. However, it lacks an explicit statement about the return value (e.g., 'Returns the created standing query'). Given no output schema, this is a minor gap.

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

Parameters3/5

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

Schema coverage is 100% with detailed parameter descriptions. The description adds minimal parameter-specific insight beyond the schema (e.g., it reiterates that required_tools must be read tools). While it provides overall context, it does not significantly enhance understanding of individual parameters.

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?

The description starts with a specific verb-resource pair ('Save a buyer-defined standing query (custom recipe)') and elaborates on what it contains (named recruiting question, prompt, read-tool list). It clearly distinguishes this create action from sibling tools by emphasizing validation against read tools and the persistence behavior.

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?

The description implies the tool is for saving a custom recipe for recurring questions, providing clear context. However, it does not explicitly state when not to use it or mention alternative tools (though siblings include other create/update tools, no direct alternative exists).

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

All Greenhouse Ops Control Plane tools Inspect

Try in Browser

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

B3.2/5.0

Disambiguation4/5

With 87 tools, there is potential for confusion among similar list tools (e.g., list_candidates vs list_applied_candidate_tags), but each tool targets a distinct entity or action, and descriptions provide enough clarity to differentiate them.

Naming Consistency5/5

All tools follow a consistent verb_noun pattern using lowercase with underscores (e.g., list_jobs, get_application, patch_candidate_profile). No mixing of styles or irregular conventions.

Tool Count1/5

At 87 tools, the count is extremely high for a single server, exceeding the 50+ threshold for extreme mismatch. While the domain is broad, the number of tools is likely to overwhelm agents and dilute focus.

Completeness3/5

The tool set covers a wide range of operations (list, get, patch, add, remove, upsert) across many entities, but lacks basic CRUD for core objects like jobs and candidates (no create_job or create_candidate), leaving notable gaps.

Resources

Need Help?