Personal crypto context layer. Provides your portfolio, investor profile, strategy notes, and trading patterns to AI agents.
Two tools with minimal schema documentation and vague descriptions. get_portfolio has a basic optional parameter but lacks output schema documentation. get_context has an optional query parameter but no schema details. Both descriptions are generic and lack specificity about return structure, error cases, or when to use each tool. No parameter type constraints (enums, ranges). No documented output schemas. No error handling guidance. Naming is acceptable (verb_noun pattern) but descriptions fall short of the 50-200 char LLM-optimized baseline.
Get the user's full crypto investor profile. Returns the holistic investor profile, the user's own strategy notes, the current portfolio snapshot, trading patterns and style analysis, and fund flow (deposits/withdrawals). All dimensions are included by default.
Get the user's current crypto portfolio. Returns holdings across all connected exchanges with asset names, amounts, USD values, and allocation percentages.
No output schemas documented for either tool. LLMs cannot plan downstream calls or extract required fields without knowing response structure.
Parameter descriptions are missing or trivial. 'asset' in get_portfolio lacks type/format constraints. 'query' in get_context is vague about expected format or when it's needed.
Tool descriptions lack specificity. get_context description is 200+ chars but vague about what 'holistic investor profile' or 'trading patterns' actually return. No guidance on when to call get_context vs get_portfolio.
No error handling guidance. Descriptions do not explain what happens if portfolio is empty, user has no connected exchanges, or query is malformed.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 46 | 2026-07-28+ | v2 |
Input schemas lack type constraints. 'asset' parameter has no enum of valid assets (BTC, ETH, etc.). 'query' has no length limits or format guidance.