Skip to main content
Glama
cirwel

UNITARES

by cirwel
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Hybrid

consult

Send a brief to get advisory model evidence for answer, critique, summarize, or generate tasks; returns policy outcome, not a governed verdict—use request_review for verdicts.

Instructions

Primary advisory model-help surface: send a brief, get back advisory model evidence, never a governed verdict — request_review produces that. effort='thorough' asks a strong model (Claude, Codex or Antigravity) from a family other than the caller's, when detectable. It needs privacy='cloud_allowed' and an operator extension a default install lacks (see list_inference_hosts); without both it fails unless allow_degraded=true, which returns a standard local answer instead. Requires a bound identity. Audited as event_type='consultation', readable by bound agents: route and keyed hashes, never text (key: record.hash_key). A success also updates your governance state. Use call_model or delegate_inference only for explicit provider, host, model or timeout control.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
briefYesQuestion or material to send for advisory model help.
effortNostandard uses the lower-overhead inference lane; thorough requests the operator-authorized strong-model lane.standard
privacyNolocal confines routing to the configured local inference service; cloud_allowed permits, but does not require, external processing.local
purposeNoDesired advisory operation. critique remains model advice, not a governed peer-review verdict.answer
agent_idNoUUID; leave unset for yourself.
response_modeNocompact returns the advisory result and policy outcome; full adds a single diagnostics object with route and inference provenance.compact
allow_degradedNoAllow thorough effort to fall back to standard local inference. This never weakens the requested privacy policy.
continuity_tokenNoSame-process rebind proof only; never a cross-process resume.
client_session_idNoBinding id for calls in this process; not a cross-process proof.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed3 schema fields changedv3.1.0
    • changedInput schema / properties / agent_id / description
      Previous value: -"UNIQUE agent identifier; optional when session-bound (auto-injected)."New value: +"UUID; leave unset for yourself."
    • changedInput schema / properties / client_session_id / description
      Previous value: -"In-session binding id from start_session()/identity(); pass it on same-process calls. Not a cross-process proof."New value: +"Binding id for calls in this process; not a cross-process proof."
    • changedInput schema / properties / continuity_token / description
      Previous value: -"Ownership proof from onboard()/identity(), for same-live-process rebinds only. Not a cross-process resume credential."New value: +"Same-process rebind proof only; never a cross-process resume."
  2. First observed

TDQS

A4.8/5.0
Behavior5/5

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

Annotations only flag readOnly=false, openWorld=true, idempotent=false; the description adds far more: the thorough lane's strong-model cross-family routing, the cloud_allowed + operator-extension prerequisite, the degraded fallback semantics, the bound-identity requirement, the audit event_type and hash-only visibility, and that success mutates governance state. This is rich disclosure beyond the annotations.

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 core purpose and the key alternative, then layers prerequisites and side effects. Dense and information-rich, though the middle sentences are tightly packed with clauses; nothing is truly wasted, but it borders on overload for a single paragraph.

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

Completeness5/5

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

No output schema, yet the description covers the return shape (advisory result + policy outcome, diagnostics with route and provenance under full), the failure modes, the identity prerequisite, and the side effects. An agent has everything needed to invoke it correctly or fall back.

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

Parameters4/5

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

Schema coverage is 100%, so baseline is 3, but the description adds real parameter meaning the schema lacks — that thorough needs privacy='cloud_allowed' plus an operator extension and otherwise fails unless allow_degraded=true, and that allow_degraded never weakens the requested privacy. It does not elaborate on purpose, response_mode, or the session/token params beyond the schema.

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?

Names a specific action and resource ('send a brief, get back advisory model evidence') and explicitly separates itself from the sibling that produces verdicts ('never a governed verdict — request_review produces that'). An agent can distinguish consult from request_review, call_model, and delegate_inference without opening any schema.

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

Usage Guidelines5/5

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

Gives explicit when-to-use (advisory model help) and when-not (governed verdicts → request_review; explicit provider/host/model/timeout control → call_model or delegate_inference). Also names the preconditions (privacy=cloud_allowed + operator extension, else allow_degraded=true) and the identity requirement.

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

Deploy Server

Other Tools