Loading...
Loading...
Found 1,645 Skills
Investment idea generation — systematically surfaces new investment opportunities by combining quantitative screening (low valuation / high momentum / improving fundamentals), thematic research (sector trends / policy catalysts), and pattern recognition (historical analogues), producing a long/short candidate list. Triggers: "投资想法", "选股灵感", "投资机会", "找股票", "发掘机会", "多头机会", "空头机会", "主题投资", "投資想法", "選股靈感", "投資機會", "找股票", "多頭機會", "空頭機會", "主題投資", "investment ideas", "stock ideas", "investment opportunities", "idea generation", "long ideas", "short ideas", "thematic investing", "stock discovery", "find me stocks", "what should I buy".
Analyst EPS estimate time series for a listed stock via Longbridge — tracks how the consensus EPS forecast (high / low / mean / median / analyst count) has changed over time, and shows actual values where reported. Answers "have analysts been raising or cutting their estimates?" Complements longbridge-consensus (current snapshot) by focusing on the historical revision trajectory. Triggers: "分析师预测历史", "EPS预测趋势", "一致预期变化", "预期上调下调", "分析师预期轨迹", "分析师EPS", "预测时间序列", "分析師預測歷史", "EPS預測趨勢", "一致預期變化", "預期上調下調", "分析師EPS", "analyst estimate history", "EPS estimate trend", "consensus revision history", "estimate time series", "analyst forecast trajectory", "EPS upgrade downgrade history".
Graham cigar-butt (NCAV / net-net) single-stock diagnostic. Combines a 100-point static cheapness score (NCAV, PE, PB, dividend yield, debt coverage, earnings stability) with a dynamic adjustment layer (industry cycle, earnings trend, insider activity, NCAV trajectory) to separate real bargains from value traps. Pulls data from Longbridge CLI/MCP first, falls back to WebSearch only for gaps, runs cross-statement reconciliation (勾稽校验) before scoring, and footnotes every figure to its source. Triggers: "格雷厄姆", "捡烟蒂", "烟蒂股", "烟蒂投资", "NCAV", "净流动资产", "清算价值", "安全边际", "价值陷阱", "深度价值", "撿煙蒂", "煙蒂股", "煙蒂投資", "淨流動資產", "清算價值", "安全邊際", "價值陷阱", "深度價值", "Graham", "cigar butt", "net-net", "liquidation value", "value trap", "margin of safety", "deep value", "Benjamin Graham".
Migrates .NET test projects from VSTest to Microsoft.Testing.Platform (MTP). Use when user asks to "migrate to MTP", "switch from VSTest", "enable Microsoft.Testing.Platform", "use MTP runner", or mentions EnableMSTestRunner, EnableNUnitRunner, or UseMicrosoftTestingPlatformRunner. USE FOR: MTP behavioral differences vs VSTest (exit code 8, zero tests discovered), --ignore-exit-code, TESTINGPLATFORM_EXITCODE_IGNORE. Supports MSTest, NUnit, xUnit.net v2 (via YTest.MTP.XUnit2), and xUnit.net v3 (native MTP). Covers runner enablement, CLI argument translation, xUnit.net v3 filter migration (--filter-class, --filter-trait, --filter-query), Directory.Build.props and global.json configuration, CI/CD pipeline updates, and MTP extension packages. DO NOT USE FOR: migrating between test frameworks (MSTest/xUnit/NUnit), xUnit.net v2 to v3 API migration, MSTest version upgrades (use migrate-mstest-* skills), TFM upgrades, or UWP/WinUI test projects.
Use when starting feature work that needs isolation from current workspace or before executing implementation plans - creates isolated git worktrees with smart directory selection and safety verification
Adversarial robustness engineering for ML/AI—evasion, poisoning, extraction, membership-inference threat models; robust training, sanitization, detectors; ASR/certified evals; lab model attacks; data-pipeline integrity; production I/O guardrails (classical ML and LLM/multimodal). Use for adversarial examples, robustness suites, poison audits, deploy guardrails—not LLM app red team (ai-redteam), governance (ai-risk-governance), safety classifier R&D (ml-research-engineer-safeguards), safeguard serving (ml-infrastructure-engineer-safeguards), privacy research (privacy-research-engineer-safeguards), AppSec pentest (penetration-tester).
Guides privacy research engineering for safeguards—PII and sensitive-data detection research, redaction and de-identification evals, memorization and extraction risk studies, privacy benchmarks and labeled corpora, logging/retention minimization for safety pipelines, and research memos on privacy–utility trade-offs for guardrail systems. Use when measuring PII detector quality, designing privacy eval suites for moderation stacks, studying training-data leakage or prompt logging risk, or recommending privacy mitigations for safeguard models—not for SOC 2/GDPR evidence automation (compliance-engineer), legal DPIA or AI policy (ai-risk-governance), harm/toxicity classifier R&D (ml-research-engineer-safeguards), production inference gateways (ml-infrastructure-engineer-safeguards), or general non-privacy research (ai-researcher).
Generate a fully working React + Vite app that explains a codebase's workflows, data types, and architecture through interactive visuals — click-to-step animated walkthroughs with auto-play, sequence diagrams, animated packet tracers, message inspectors that toggle between named-field view and raw JSON, and collapsible code peeks with file:line citations. Splits the repo into 4–6 domain clusters and dispatches one content agent per cluster to write the pages in parallel. The skill bundles its own reference pages (under references/examples/) so it works in any repo. Use this skill whenever the user asks for interactive docs, animated explainers, an "agent team" for docs, one page per domain, wants to visualize a system's request flow or wire protocol, or any visual documentation site. Requires CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1 in .claude/settings.json.
Visualize a specific transformer decoder layer from an AutoDeploy FX graph text dump as a hierarchical DOT/PNG diagram. Optionally annotate nodes with actual GPU kernel names and durations from an nsys trace. Use when the user wants to visualize, inspect, or debug a layer in an AutoDeploy model graph dump. Triggers on: "visualize layer", "show layer", "graph of layer", "layer visualization", "dump graph layer". Assumes graph dumps already exist in a directory (produced by AD_DUMP_GRAPHS_DIR).
Use when an ops leader (Director of CX, Head of Support, VP Ops, Head of BizOps, Head of IT ops, Head of Finance ops) is sizing ops capacity, building a headcount plan, modeling utilization risk, planning Q3 capacity or annual support capacity, or designing CS coverage — and needs Erlang-C queueing math, P90 demand sizing, shrinkage-adjusted FTE, manager-trigger thresholds, and a quarterly hiring sequence with ramp + attrition. Apply when sustained team utilization is above 80% or when the team is growing >50% in 12 months. Run before committing the headcount budget. This is NOT engineering capacity (see vpe-advisor for DORA + cycle time) and NOT strategic 3-year workforce planning (see chro-advisor).
Use for Bun file I/O: Bun.file, Bun.write, streams, directories, glob patterns, metadata.
Convert Markdown documents into beautiful, comfortable, well-structured, directly openable HTML long-form reports. Suitable for converting .md files to HTML, standardizing the layout of research/industry/survey documents, generating single-file web reports for offline sharing, embedding or verifying local images, fixing misaligned columns caused by Markdown table separators, or optimizing reading margins, image presentation, directory navigation, table responsiveness, and print styles of existing HTML reports.