A Clawdbot skill that gives your agent native access to DWLF — a market analysis platform for crypto and stocks.

**Judgement:** High-value, feature-rich skill for agents that need deep, DWLF-native market analysis and trading workflows. Best suited to power users of DWLF, trading assistants, and research agents that must work directly with structured market, signal, and portfolio data. **Key strengths** - Broad coverage of the DWLF platform: market data, indicators, S/R, trendlines, events, signals, strategies, backtesting, portfolios, trades, watchlists, annotations, plans, and position sizing. - Clear, AI-oriented endpoints (`/ai/dashboard`, `/ai/symbol-brief/{symbol}`, `/ai/strategy-performance`) to obtain rich, pre-aggregated views instead of juggling many low-level calls. - Strong support for advanced technical concepts (EMA, RSI, DSS, MACD, SMC, order blocks, FVGs, BOS/ChoCH, etc.) plus auto-detected structures (support/resistance, trendlines, patterns). - Full event/strategy lifecycle exposed (create → compile → activate/deactivate symbols → evaluate), enabling sophisticated automation of alerts and systematic strategies. - Integrated trading workflow utilities: position sizing, trade plans, annotations, portfolio & trade tracking, watchlists, and AI-accessible DWLF Academy content for explanations. **Notable implementation details / pitfalls** - **Symbol formats are strict:** - Crypto must be like `BTC-USD`, `ETH-USD`, `SOL-USD` (with `-USD` suffix). - Stocks/ETFs use bare tickers (`TSLA`, `NVDA`, `META`, etc.). - Forex uses pairs like `GBP-USD`, `EUR-USD`. Agents must normalize casual user inputs (e.g., "BTC" → `BTC-USD`). - **Backtests are asynchronous:** `POST /backtests` only triggers the run. Agents must poll `GET /backtests/{requestId}` until status is `"completed"`, then fetch results via `GET /backtests/{requestId}/results`. This adds complexity and latency. - **Events & strategies require compilation and activation:** - After `POST /custom-events` or `POST /visual-strategies`, agents must 1) `POST /custom-events/{id}/compile` or `POST /visual-strategies/{id}/compile`, and 2) Explicitly activate symbols: `POST /custom-event-symbols/:eventId/enable-all` or `POST /strategy-symbols/:strategyId/enable-all` with `{ "symbols": ["BTC-USD", ...] }`. - Any later `PUT` update also requires recompilation; if skipped, changes have no effect. - Without activation, events/strategies will not fire or generate signals, which can confuse both user and agent. - **Use AI summaries first:** For generic market or account questions (“How’s BTC?”, “What’s going on?”, “How is my account doing?”), `/ai/dashboard` and `/ai/symbol-brief/{symbol}` should be preferred to manual compositions of market, S/R, indicators, and events. - **Academy integration for explanations:** When users ask “How does DSS work in DWLF?” or “What is SMC here?”, the recommended pattern is to call the academy tools (`dwlf_list_academy_tracks`, `dwlf_search_academy`, `dwlf_get_academy_lesson`) rather than invent explanations. This can materially improve factual accuracy. **Risks / limitations** - **Financial domain risk:** The tool exposes data, signals, and backtests in a high-risk domain (trading). Agents must avoid turning this into implicit financial advice and should communicate uncertainty and limitations of backtests, indicators, and signals. - **Over-reliance on signals/backtests:** Users may overinterpret strategy performance or recent signals as guarantees. Agents should emphasize that backtests are historical, can be overfit, and may not generalize; signals are probabilistic and can fail. - **External dependency & uptime:** Performance and reliability are bounded by DWLF’s API stability, latency, and rate limits, which are not detailed here. Agents should be prepared to handle timeouts, partial results, and transient errors. - **Complex workflows:** Advanced flows (e.g., building and iterating on custom strategies, coordinating async backtests, ensuring proper symbol activation) require careful orchestration. Poor tool usage can lead to silent failure modes (no signals, stale strategies, incomplete backtests). **Recommended scenarios** - Building a DWLF-native trading assistant that can: - Answer “how’s the market?” or “how’s BTC/TSLA?” using AI summary endpoints. - Provide structured technical analysis and highlight recent events, signals, and key S/R levels. - Help design and iterate on rule-based strategies and custom events, then run backtests and summarize performance. - Manage user portfolios, watchlists, and trade plans, and calculate position sizes given risk constraints. - Educational or coaching agents that leverage DWLF Academy content to explain indicators, strategies, and SMC concepts while grounding explanations directly in platform-native material. **Less ideal scenarios** - Applications that require ultra-low-latency or HFT-style execution; this tool is oriented toward analysis, not millisecond-level trading. - Simple, non-DWLF users who only need generic price lookups—this skill is overkill if no DWLF workflows are involved.