Explore skills

Create or update the data flow architecture document (docs/pages/data-flow.md).

[PM Twin] Your digital peer product manager — assess project state, fill gaps, recommend next action.

Update stale documentation and comments to match current code

[PM Twin] Sprint-boundary verification gate — runs every feature in the sprint through qa-feature, runs the sprint's end-to-end journey flows, aggregates evidence, and gates the next sprint.

Remove dead code, fix duplication, and optimize imports in modified files

Analyzes changes and generates Conventional Commit messages

Configure deployment script and run it, verifying it succeeds

Decodes market hotspots, thematic tracks, and industry events, focusing on recent core events, opportunities, and risks. It automatically extracts thematic keywords and generates structured hotspot event decoding reports using a financial data interface.

Manages AI IDE rule files across Claude Code, Cursor, Windsurf, Trae, and GitHub Copilot, supporting the creation, modification, and deletion of rules, project initialization, and cross-IDE synchronization.

Implements and debugs Binance Spot REST /api/* calls, including /api/v3 market data, signed account/orders, and security mechanisms like HMAC/RSA/Ed25519 signing. Handles recvWindow, filters, rate limits, errors, and supports Testnet.

Work with test skeletons: transcribe test names from design documents into code stubs, add test skeletons to design documents, or draft new design documents including them. Use this skill when "test skeleton", "test stubs", "transcribe tests" are mentioned, or for tasks creating placeholder tests from a plan.

Self-improving knowledge wiki with a vault of raw sources. Triggers on terms like 'curiosity engine', 'wiki', 'vault', 'knowledge base', 'ingest', 'iterate', 'refine', 'improve', 'evolve', 'curator', 'lint', or when adding sources, querying knowledge, or improving the wiki.