Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
BOARDROOM_ROOTNoPath to your full protocol files directory~/.ai/boardroom
BOARDROOM_TRUST_PATHNoPath to trust oracle data file~/.boardroom/trust-oracle.json

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Capabilities

Features and capabilities supported by this server

Protocol revision2025-11-25

CapabilityDetails
tools
{
  "listChanged": true
}

Tools

Functions exposed to the LLM to take actions

NameDescription
analyzeA

Run a Boardroom consultation. Routes your question to relevant advisors, loads their philosophies and decision criteria, searches institutional memory for precedents, and provides a structured analysis with mandatory tension between opposing viewpoints. Demo mode includes 3 named advisors; full protocol files unlock 450+ advisors across 38 councils.

check_governanceA

Classify a task and determine which governance advisors should review it. Returns the decision type, selected advisors, risk level, and whether constitutional constraints apply. Fast classification without running the full session.

query_intelligenceA

Search the Boardroom LEDGER (persistent decision memory) and Wisdom Codex for relevant precedents, past decisions, and distilled insights. Returns keyword-matched results with timestamps and excerpts. The LEDGER grows each time you use report_outcome.

trust_lookupA

Look up the trust profile for any entity (AI agent, tool, vendor, platform). Returns a 6-dimension trust vector (reliability, honesty, follow-through, outcome quality, stability, risk profile), composite score, and recommendation (trust/verify/caution/avoid). New entities return a default "unknown" profile — use report_outcome to build trust data over time.

report_outcomeA

Report the outcome of a decision for the Boardroom learning system. Records what happened, whether the original recommendation was followed, and what was learned. Feeds the Knowledge Flywheel and updates the Trust Oracle if an entity is specified. Returns a warning if the outcome could not be persisted to disk.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

TDQS

A3.9/5.0

Scored across 5 tools

Disambiguation4/5

Each tool targets a distinct action: classification, search, trust lookup, outcome reporting, and full consultation. The only minor overlap is between query_intelligence and analyze, as both access institutional memory, but their purposes are clearly differentiated by analysis' routing to advisors.

Naming Consistency3/5

Tool names mix verb_noun patterns (check_governance, query_intelligence, report_outcome) with a noun_verb compound (trust_lookup) and a single verb (analyze). This is readable but not fully consistent, making the pattern less predictable than it could be.

Tool Count5/5

Five tools is well-scoped for a governance/consultation server, covering classification, memory search, trust evaluation, outcome feedback, and full analysis without redundancy. Each tool earns its place in the workflow.

Completeness4/5

The core lifecycle is covered: classify, analyze, query, report, and trust lookup are present, with feedback loops for learning. Minor gaps exist (e.g., no direct tool for editing governance protocols), but agents can work around this with the existing tools.

Maintenance

ActivityInactive
ResponsivenessNo issues