trader-pro

External API evaluation and MCP server wrapper generation. Load when integrating a REST/GraphQL API as a workspace tool

Capability-based provider system patterns. Load when creating providers or wiring module-provider integration

Full-stack WebSocket subscription development. Load when adding WS routes, topic services, or debugging WS data flows

Systematic Python type error resolution (mypy/pyright). Load when fixing backend type check failures

Type-first Python with Pydantic, Protocol, and discriminated unions. Load when writing models or enforcing typing

Systematic type error resolution for Python (mypy/pyright) and TypeScript (vue-tsc). Use when fixing type errors, resolving type check failures, or optimizing type imports.

Terminal command safety, delegation routing, and guarded execution. Load when running shell commands or delegating to Bash subagents

Non-destructive investigation constraints. Load when analyzing, debugging, reviewing, or performing RCA without edits

Structured user input gathering via ask_questions — decision flowchart, component patterns, inference heuristics. Load when facing vague instructions or designing prompts that gather user input

VS Code MCP tool routing, diagnostics-driven edit loops, and IDE sub-tool reference. Load when choosing between native and MCP workspace tools or using VS Code IDE features

WCAG 2.1 AA patterns for inclusive UI. Load when building components needing a11y, ARIA, or keyboard nav

Pytest patterns for trader-pro modular backend. Load when writing, debugging, or analyzing backend test coverage

Post-implementation review methodology with severity framework. Load when reviewing code, PRs, or auditing quality

Hypothesis-driven debugging methodology. Use when investigating root cause, diagnosing failures, issues or problems.

Solution selection and adversarial validation. Load when proposing solutions or evaluating architecture decisions

Systematic TypeScript/Vue type error resolution (vue-tsc/ESLint). Load when fixing frontend type check failures

Distinctive, production-grade visual design for frontend interfaces. Load when building UI needing polished aesthetics

Vitest/Vue 3 testing patterns for trader-pro frontend. Load when writing or debugging frontend tests

Full runtime environment inventory — platform, tools, services, paths. Load when checking tooling or capabilities

Shared test methodology for backend and frontend. Load when planning test suites or evaluating coverage value

TradingView Trading Terminal API patterns and type conventions. Load when implementing broker/datafeed or TV integration

TradingView bundle maintenance and debugging. Load when debugging TV bundle issues or patching obfuscated JS

Contract-first TypeScript types for generated clients and mapper isolation. Load when writing mappers or API types

UX design thinking for friction reduction and cognitive load. Load when building interactive UI or auditing user flows

Vue 3 Composition API + TypeScript patterns. Load when building components, composables, Pinia state, or reviewing Vue

Create VS Code custom agents (.agent.md files) with proper YAML frontmatter, constraint craft, skill integration, and handoff design. Use when creating agents, configuring subagents, designing handoff chains, or writing effective constraints.

Curated directory of online resources for discovering prompts, agents, skills, and MCP tools. Use when searching for reusable AI coding assets, evaluating skill/tool marketplaces, installing community skills, finding MCP servers, or recommending resources to users. Covers official vendor sources, community marketplaces, and CLI tooling with assessment criteria.

VS Code prompt file (.prompt.md) format specification and design principles. Use when creating prompt files, designing reusable prompts, or reviewing prompt templates.

Category glossary for frontend. Contains 5 skills: `accessibility`, `fix-frontend-type-errors`, `typescript-contract-...

Reasoning guardrails for implementation agents — prevents design-spiral, enforces materialization checkpoints, and steers thinking toward actionable coding steps. Use when planning implementation, entering complex reasoning chains, or detecting reasoning drift away from concrete code changes.

Break complex problems into tractable sub-problems with clear interfaces, dependencies, and validation. Use when a problem has multiple interacting dimensions, compounding uncertainty, or when direct attack exceeds working memory.

Agent delegation, subagent routing heuristics, and invocation quality standards. Use when deciding to delegate, handoff, or spawn subagents — ensures comprehensive context and accurate output descriptions for every subagent invocation.

Python/pytest backend testing patterns for the trader-pro modular architecture. Use when writing backend tests, analyzing coverage, selecting fixtures, or debugging test failures. Complements test-strategy skill with pytest-specific patterns.

Post-implementation code review methodology with severity framework and project-specific rules. Use when reviewing code changes, auditing pull requests, analyzing code quality, or performing security review.

Vitest/Vue 3 frontend testing patterns for the trader-pro architecture. Use when writing frontend tests, selecting test patterns, mocking services/WebSockets, or analyzing frontend coverage. Complements test-strategy skill with Vitest-specific patterns.

Large data volume handling — detection triggers, delegation protocols, and convergence gates for files, diffs, search results, and command output. Use when processing >200-line files, large git diffs, bulk search results, or extended investigations without convergence.

Validation workflow for IA stack artifacts (agents, subagents, prompts, skills). Use when validating existing artifacts, checking boundary compliance, classifying artifact types, or generating structured validation reports with severity-ranked findings.

Request immunity patterns for user-facing agents — tier selection (T1/T2/T3), Phase 0 injection templates, and RV gate definitions. Use when creating user-facing agents, injecting request validation, or checking immunity gate compliance.

Change impact analysis and incremental modification protocol for interconnected agentic design stacks. Use when modifying, renaming, restructuring, or removing existing IA assets (agents, skills, subagents, prompts, templates). Prevents big-bang changes that destabilize learned system behaviors.

Shared test methodology patterns for backend and frontend test agents. Use when planning test suites, decomposing test surfaces, evaluating coverage value, or structuring test validation with post-action reflexion.