idf-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| IDF_DOMAIN | No | The domain name to use. Corresponds to --domain flag. | booking |
| IDF_SERVER | No | The URL of the Fold runtime (IDF host). Corresponds to --server flag. | http://localhost:3001 |
| IDF_BOOTSTRAP | No | Set to 0 to skip bootstrap (domain already loaded). Corresponds to --no-bootstrap flag. | 1 |
| IDF_AGENT_EMAIL | No | Email for the agent identity. Corresponds to --agent-email flag. | mcp-agent@local |
| IDF_ONTOLOGY_PATH | No | Path to the IDF ontology directory. Corresponds to --ontology-path flag. | ./src/domains/<domain> |
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
| Capability | Details |
|---|---|
| tools | {} |
| resources | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| agent_propose_rebalanceB | Robo-advisor proposes a portfolio rebalance with a confidence score and rationale. |
| agent_execute_preapproved_orderA | Robo-advisor executes a market order. Subject to AgentPreapproval guard (active / notExpired / maxOrderAmount / allowedAssetTypes / dailyLimit / dailySum). May fail on (domain invariants):
|
| agent_flag_anomalyA | Robo-advisor raises a portfolio-level alert with severity. |
| agent_fetch_market_signalA | Robo-advisor records a market signal (price / volume / news) for an asset. |
| agent_recompute_risk_scoreA | Robo-advisor recomputes the risk profile for a portfolio (or all portfolios if portfolioId omitted). |
| agent_generate_reportA | Robo-advisor generates a portfolio-level performance / risk report as a recommendation. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| users | Filtered collection of User — поля: id, name, email |
| portfolios | Filtered collection of Portfolio — поля: id, userId, name, baseCurrency, riskProfile, targetStocks, targetBonds, targetCrypto, targetExotic, totalValue, pnl |
| positions | Filtered collection of Position — поля: id, portfolioId, userId, assetId, quantity, avgPrice, currentPrice, unrealizedPnL, stopLoss, takeProfit |
| assets | Filtered collection of Asset — scope: all |
| riskProfiles | Filtered collection of RiskProfile — поля: id, userId, level, computedScore, updatedAt |
| recommendations | Filtered collection of Recommendation — поля: id, userId, source, type, status, confidence, createdAt |
| alerts | Filtered collection of Alert — поля: id, userId, severity, message, triggeredAt |
| marketSignals | Filtered collection of MarketSignal — scope: all |
| agentPreapprovals | Filtered collection of AgentPreapproval — поля: id, userId, active, maxOrderAmount, allowedAssetTypes, allowedPortfolioIds, dailyLimit, expiresAt |
TDQS
Scored across 6 tools
Each tool targets a distinct action in the robo-advisor workflow: order execution, signal fetching, anomaly flagging, report generation, rebalance proposal, and risk score recomputation. No two tools have overlapping purposes.
All tools follow a consistent verb_noun pattern with the 'agent_' prefix, using imperative verbs like 'execute', 'fetch', 'flag', 'generate', 'propose', and 'recompute' followed by clear nouns.
With 6 tools, the set is well-scoped for a robo-advisor, covering core operations without unnecessary bloat or missing essentials.
The tools cover key robo-advisor functions, but missing a tool for viewing portfolio details or managing preapproval settings could cause minor gaps for agents.