Loading ToolBench
MCP server for portfolio management, risk analysis, and investment decision tracking integrated with PaytmMoney and INDmoney brokers
21 tools with consistent naming (verb_noun pattern: get_*, set_*, add_*, log_*) and descriptions present. However, most descriptions are generic (50-100 chars) and lack actionable context for LLM selection. Input schemas visible for all tools, but many lack parameter descriptions or constraints. No output schemas documented. Error handling and recovery guidance absent. Tool composition is reasonable (single responsibility mostly observed), but parameter design could be more LLM-friendly (e.g., symbol as free-form string vs. constrained enum).
Add a new investment thesis entry for a position with rationale and target metrics
Progress toward allocation targets across dimensions (sector, asset class, region)
Behavioral finance analysis including disposition effect, averaging down patterns, and win/loss asymmetry
Portfolio performance benchmarked against market indices (Nifty 50, Sensex, etc.)
Comprehensive summary of all Compass features (goals, allocation targets, milestones) in one call
Real-time portfolio dashboard with holdings, P&L, allocation breakdown, and risk metrics
No output schemas documented for any tool. LLMs cannot plan downstream calls or extract required fields without knowing response structure.
Parameter descriptions missing or minimal for read-only tools (get_*). E.g., get_thesis_history accepts 'symbol' and 'limit' but lacks format/range guidance. LLMs cannot validate inputs without constraints.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 61 | <=2025-11-25 | v2 |
Investment decision log with tracked calls, outcomes, and performance metrics
Dividend history and upcoming dividend payments for portfolio holdings
Progress toward investment goals with current status and gap analysis
Progress toward financial milestones with target dates and current status
Risk analysis including volatility stops, sector concentration, and portfolio risk metrics
System status including broker connection status, last sync time, and data freshness
Tax optimization suggestions including loss harvesting opportunities and tax-loss carryforward analysis
Investment thesis history with entries, rationale, and outcomes for tracked positions
Current risk threshold settings for volatility stops and position limits
Historical portfolio trend data including daily snapshots and performance metrics
Volatility-based stop-loss recommendations for current holdings
UI action to highlight a specific holding in the frontend dashboard
Log a new investment decision with success criteria and tracking parameters
Update the status of a logged decision (accepted, dismissed, etc.)
Set or update a risk threshold (volatility stop or position limit)
Tool descriptions are generic and under 100 chars. E.g., 'Real-time portfolio dashboard with holdings, P&L, allocation breakdown, and risk metrics' does not explain WHEN to call this vs. get_risk or get_allocation_progress. LLMs struggle to disambiguate.
No error handling or recovery guidance. Tools do not document what happens on invalid symbol, missing data, or API failures. LLMs cannot self-correct or retry intelligently.
Parameter 'symbol' is free-form string across all tools. No enum, pattern, or validation guidance. LLMs may pass invalid symbols (e.g., 'AAPL' vs. 'RELIANCE') without feedback.