Skip to main content
Glama
antonbabenko

Deliberation

by antonbabenko
OverviewSchemaRelated ServersScoreDiscussions
JavaScript
Remote

consensus

Read-only

Run multi-round consensus convergence with a provider arbiter to settle questions, or synthesize open-ended queries in one pass.

Instructions

Run the FULL multi-round consensus convergence loop server-side with a provider arbiter (blind pass + peer fan-out -> adjudicate -> revise) and return the converged verdict. Default depth is consensus.maxRounds (config, default 5); pass maxRounds to override. Pass synthesizeAlways:true for a SINGLE arbiter synthesis pass instead of the loop (best for open questions, not plan convergence): it returns a free-text synthesis and maxRounds is ignored. Configure the arbiter via consensus.arbiter - a concrete provider/openrouter alias runs server-side; host mode returns the opinions for YOU to synthesize. Advisory; pass expert to apply a persona. Calls external providers (keys/CLI; rate limits apply); returns a text-wrapped JSON envelope (split verdict/synthesis, loop fields nullable) and persists a session record only when sessions.persist is enabled (default off). NOTE (Claude Code): use the /consensus slash command for the transcript-visible host-arbiter loop (it drives consensus-step); this tool is the provider-arbiter path for any host.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
cwdNoWorking directory the provider runs in (used to resolve relative file refs). Defaults to the server process directory.
filesNoOptional attachments for providers that read files (Grok/OpenRouter; inlined as context for Codex/Gemini). Each item is EXACTLY ONE of path/dir/file_id/file_url.
expertNoOptional persona: architect, plan-reviewer, scope-analyst, code-reviewer, security-analyst, researcher, or debugger. On a named expert tool the tool's own persona wins and this is ignored.
promptYesThe question or task for the provider(s)/expert.
maxRoundsNoOverride consensus.maxRounds for this call (loop mode only; ignored when synthesizeAlways is true). Clamped to 50.
reasoningEffortNoReasoning depth where the provider supports it (Grok, OpenRouter): low, medium, high, or none. CLI providers (Codex, Gemini) ignore it.
synthesizeAlwaysNoRun ONE arbiter synthesis pass instead of the convergence loop. Returns a free-text `synthesis` (verdict/converged/confidence are null, rounds is 1). Best for open questions.
developerInstructionsNoOptional system/developer instructions injected verbatim; overrides the built-in persona for `expert`.

Tool Definition Quality

A4.8/5.0
Behavior5/5

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

Discloses important behavioral traits: calls external providers with rate limits, returns a text-wrapped JSON envelope with split verdict/synthesis, and persists session only when enabled. No contradiction with annotations (readOnlyHint, destructiveHint, openWorldHint).

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 dense with information and front-loaded with the core purpose. Every sentence provides necessary context. The NOTE at the end is somewhat audience-specific but does not detract significantly. Could be slightly trimmed, but overall efficient for the tool's complexity.

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?

Given 8 parameters, no output schema, and moderate complexity, the description covers purpose, usage, parameters, behavior, and return format (JSON envelope). It is self-contained and eliminates ambiguity for the agent invoking the tool.

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% with descriptions for all 8 parameters. The description adds value beyond schema by explaining default behavior (e.g., maxRounds default from config, synthesizeAlways overrides loop, expert persona overriding, file delivery modes). Reduces cognitive load for the agent.

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 clearly states it runs a 'multi-round consensus convergence loop server-side' and returns a converged verdict. It distinguishes from sibling tool 'consensus-step' via the NOTE about using the /consensus slash command for the host-arbiter loop, and from other single-provider tools by the loop mechanism.

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?

Explicitly describes when to use the full loop vs. the single synthesis pass (synthesizeAlways), and advises on open questions vs. plan convergence. Also mentions the alternative /consensus slash command for host-arbiter loop, providing clear selection criteria.

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

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/antonbabenko/deliberation'

If you have feedback or need assistance with the MCP directory API, please join our Discord server