The BAMM MCP server has significant definition quality gaps. Tool names follow verb_noun convention (ADD_COLLATERAL, BORROW, GET_POSITIONS, etc.) which is good, but descriptions are terse and lack context about when to use each tool or what they return. Parameter descriptions are present but minimal (5-15 words). Most critically, NO input or output schemas are visible in the provided source code, only parameter names and types are mentioned in the evaluation brief. The absence of explicit JSON Schema definitions in the tool registrations is a major red flag. Tool files reference (add-collateral.ts, borrow.ts, etc.) are listed but source code not provided, so actual schema quality cannot be verified. This forces conservative scoring. No tool descriptions mention state mutation clearly, error recovery, or expected outputs. Parameter descriptions lack format constraints, ranges, or examples of valid values.
Add collateral tokens to a BAMM position
Borrow tokens from a BAMM pool by providing collateral
Get all active BAMM positions for the connected wallet
Lend Fraxswap LP tokens to a BAMM pool to earn interest
Get statistics for all available BAMM pools
Remove collateral tokens from a BAMM position
Repay borrowed tokens to a BAMM pool
No input/output schemas visible in source code. Tool definitions appear inferred from brief rather than explicitly visible in provided code. Schema files not provided.
Tool descriptions are extremely short (10-40 chars) and lack context about WHAT the tool does, WHEN to use it, or what it returns.
Parameter descriptions are minimal (e.g., 'The BAMM pool address', 'The amount to add in normal decimal form') and lack constraints, ranges, or format specifications. LLM cannot infer valid values from these terse annotations.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | F | 14 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 36 | - | v1 |
No documentation of output schemas or return types. LLM cannot plan downstream tool calls or extract chaining IDs (e.g., if ADD_COLLATERAL succeeds, what fields does it return?)
No error handling documentation visible. Tools that perform writes (ADD_COLLATERAL, BORROW, LEND, REMOVE_COLLATERAL, REPAY) should document failure modes and recovery guidance.
Optional parameters (collateralToken, collateralTokenSymbol, borrowToken, borrowTokenSymbol) lack clarity on mutual exclusivity or when to use name vs address. Should document dependency hints.
No dry-run or confirmation mechanism documented for irreversible write operations (BORROW, LEND, REPAY). Agents should have a safety gate before executing state-mutating calls.