Content
37%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides a broad catalog of example commands for crypto trading via Bankr, but critically lacks implementation details (how does Claude actually make API calls?), workflow safety measures for financial operations, and error handling guidance. The examples are illustrative but the skill reads more like user-facing documentation than an actionable guide for Claude to execute trades programmatically.
Suggestions
Add executable code showing how Claude should make HTTP requests to the Bankr API (endpoint URL, request format, authentication header, response parsing) rather than just listing slash commands.
Add a critical workflow with validation checkpoints for trades: check balance → preview trade → confirm → execute → verify status → handle errors. Financial operations without confirmation/validation steps are dangerous.
Clarify the execution model: are these slash commands for a specific MCP tool, CLI commands, or API calls? Claude needs to know the exact mechanism to use.
Move the extensive example categories (NFTs, Polymarket, Leverage, Token Deployment, etc.) to a separate reference file and keep only 3-4 representative examples in the main skill.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is mostly efficient but includes some content that could be tightened. The supported chains table and tips section add marginal value, and the sheer number of examples is somewhat redundant—Claude could infer many of these patterns from a few representative examples. | 2 / 3 |
Actionability | The skill provides concrete command examples but lacks the actual API implementation details—there's no executable code showing how the /bankr command works under the hood (HTTP requests, response handling, job polling). It's unclear whether these are slash commands in a specific tool, CLI commands, or natural language prompts to an API endpoint. | 2 / 3 |
Workflow Clarity | For a trading skill involving real financial transactions (destructive/irreversible operations), there are no validation checkpoints, confirmation steps, error handling, or feedback loops. The status/cancel commands are mentioned but not integrated into a workflow showing how to verify a trade before it executes or handle failures. | 1 / 3 |
Progressive Disclosure | The content is organized into clear sections with headers, which is good. However, the extensive list of examples across 10+ categories makes this somewhat monolithic—the many example categories could be split into a separate EXAMPLES.md or COMMANDS.md reference file, keeping the main skill leaner. | 2 / 3 |
Total | 7 / 12 Passed |