dspy-bootstrap-rs
Use when basic BootstrapFewShot is not enough and you want to search over multiple candidate demo sets — better results at the cost of more LM calls. Common…
Find an Agent Skill for your task. Read the guide and copy the install command.
Updated 11-17 00:07
53,896 results
Use when basic BootstrapFewShot is not enough and you want to search over multiple candidate demo sets — better results at the cost of more LM calls. Common…
Use when you want few-shot demos that are dynamically selected per input based on similarity — better than fixed demos when inputs vary widely. Common…
The English introduction is being prepared.
Monitoring high-impact economic calendar (NFP, US CPI, FOMC Rate, BI-Rate). Implementing the Pre-News Shield strategy, calculating the deviation of the Actual vs… surprise
Comprehensive multi-method technical analysis. Calculates classic indicators (EMA, RSI, MACD, Bollinger), Price Action, Supply & Demand, and Smart liquidity detection…
Running Google Research's Time Series Foundation Model (TimesFM) AI model to predict future price trajectories in a probabilistic zero-shot manner…
Multi search engine integration with 17 engines (8 CN + 9 Global). Supports advanced search operators, time filters, site search, privacy engines, and…
Multi search engine integration with 17 engines (8 CN + 9 Global). Supports advanced search operators, time filters, site search, privacy engines, and…
Decide which UE5 class owns each piece of runtime gameplay state by matching it to the lifetime the state must survive — Pawn, Controller, PlayerState,…
Consolidate a fact that's scattered across places into one canonical source and replace the rest with references — mutating. Use when the user asks to…
Own canonical docs and behavior specs; keep them synced with code.
Maintain LobeHub's model-backed alint rule set in packages/alint: author a rule, calibrate its false positives against real source, decide its severity (warn…
Chain IPRoyal static residential IPs into Clash Verge through RayLink encrypted node chaining, so AI websites use a fixed exit that is verifiable and rollback-capable. Use for "add IPRoyal to RayLink", "IPRoyal direct connection fails", "chain IPRoyal behind RayLink".
Generate Past Build project JSON for portfolio import. Analyzes codebase to extract project name, description, tech stack, and structure for the Past Build…
TGravity Work asset card Skill. Used to generate, update, and check creator profile cards, brand lead cards, deal advancement cards, deal retrospective cards, and decision request cards, precipitating key objects from daily reports or verbal briefs into handoff-ready Markdown assets. Trigger methods: generate asset card, update asset card, creator card, creator tiering, brand lead, brand card, deal advancement, deal status, deal retrospective, decision request, generate decision request card, save...
TGravity Work daily report Skill. Used to organize employees' natural speech, voice-to-text, chat logs, and brand/influencer/commercial order progress into a Markdown work daily report, automatically distinguishing today's completed items, commercial order progress, payment collection amounts, risk blockers, tomorrow's plan, and matters requiring Lulu's judgment. Trigger methods: /tgravity-daily, TGravity daily report, work daily report, generate today's daily report, save today…
TGravity Work pre-action review Skill for major projects. For when users are about to invest a large amount of time, money, and energy into a project, a course, a major purchase, a long-term commitment, a business transformation, or a direction adjustment, or when they are repeatedly agonizing, when everyone else is doing it but they cannot clearly say why, or when they are halfway through and feel they have gone off track, use the 60/80/120 delivery levels and 11...
Audit any live Shopify store or product page for Google Merchant Center Misrepresentation policy risks by crawling public pages. Use when a merchant was…
Choose the right product experiment type: superiority, non-inferiority, equivalence, A/B/n, or holdback-backed validation. Use when deciding what kind of A/B…
CX workflow — project blueprint discussion. Triggered when the user mentions "blueprint", "overall planning", "project scope", "scope", "project discussion", "feature proposal". Discuss the project or feature proposal through multi-turn conversation, save the results locally `开发文档/CX工作流/功能/{功能标题}/范围.md`, optionally sync to a GitHub Issue (based on…
CX workflow — progress viewing. Read project-level configuration and status files, and display the current feature, current task, blocking reasons, and recent fix records.
Scrapes Danish job sites for new positions matching your profile. Deduplicates across runs. Triggers on: job scrape, find jobs, search jobs, new jobs, job…
Idea group chat room — Analogist role. Invoked by the idea-team main orchestrator, or triggered when the user alone says "make an analogy", "is there anything in other industries", "yes-and expansion", "what does X remind you of", "cross-domain inspiration". **Specializes in cross-domain analogies + yes-and expansion — does not judge, does not nitpick, does not require factual evidence**. Do NOT use when the user wants data (use…
Idea Group Chat Room — Devil's Advocate role. Invoked by the idea-team main orchestrator, or triggered when the user individually says "opposing opinion", "poke holes in this idea", "why will it fail", "find loopholes / counterexamples", "devil's advocate". **Specializes in poking holes, finding hidden assumptions, and giving counterexamples — no comforting, no "maybe it could be like this", no wholesale rejection**. Do NOT use when…