Loading...
Loading...
Found 69 Skills
Detects fabricated content, false citations, and unverifiable claims in agent outputs. Uses source verification and consistency checking. Activate on 'detect hallucination', 'fact check', 'verify claims', 'check accuracy', 'find fabrications'. NOT for validation (use dag-output-validator) or confidence scoring (use dag-confidence-scorer).
Commuting Plan/Route Planning (point-to-point or multi-stop routing). For user queries such as 'How to get there/How to commute/Commuting plan/Subway and bus transfer/How long and how much for a taxi/From hotel to scenic spot or venue', etc. Default to real-time verification and provide source links and query time (local time); do not fabricate precise figures; Calendar entry requires explicit user confirmation.
Weather/current-conditions verification workflow. Use when user asks for weather, temperature, forecast, rain/snow/wind, typhoon alerts, air quality, or other time-sensitive local conditions. Requires location + source links + local update/publish time; NEVER fabricate.
Expert knowledge for Azure Operator Nexus development including troubleshooting, best practices, decision making, architecture & design patterns, limits & quotas, security, configuration, and deployment. Use when managing Nexus fabric BGP/VRF, ACL/QoS, NAKS clusters, near-edge storage design, or Nexus upgrades, and other Azure Operator Nexus related development tasks. Not for Azure Network Function Manager (use azure-network-function-manager), Azure Networking (use azure-networking), Azure Virtual Network Manager (use azure-virtual-network-manager), Azure Virtual WAN (use azure-virtual-wan).
Apply Benford's Law to detect anomalies in numerical datasets by analyzing first-digit frequency distributions. Use this skill when the user needs to audit financial data for fraud indicators, validate data integrity, or detect fabricated numbers — even if they say 'data manipulation detection', 'first digit test', or 'accounting fraud screening'.
High-speed rail / train lookup workflow (China-focused). Use for timetables, ticket availability, delays, cancellations, or “today/latest” rail updates. Prefer official sources; always include source link + local update time; NEVER fabricate.
Restaurant recommendations / What to eat / Where to eat / Coffee, bars, desserts, late-night snacks. For users asking questions like "What's good to eat nearby? / Restaurant recommendations / What's the average cost per person? / What time do you close? / Do I need a reservation? / Suitable for families or treating guests?" etc. By default, provide a "query time snapshot" (local time) and source links; do not fabricate restaurant names, addresses, business hours, average cost per person, or ratings; if information is insufficient, first ask 2-3 key questions.
SEO intelligence toolkit covering the full lifecycle via live web data: keyword research, rank tracking, site audits, content gap analysis, competitor keyword reverse-engineering, AI visibility across five platforms (ChatGPT, Perplexity, Google AI, Gemini, Grok), and GitHub repo SEO. Crawls real sites and SERPs via Nimble CLI — no fabricated metrics. Triggers: "SEO", "keywords", "rank tracker", "site audit", "content gap", "competitor keywords", "AI visibility", "GitHub SEO", "SERP analysis", "keyword research", "technical SEO", "keyword difficulty", "topic clusters", "ranking delta", "on-page SEO", "AI citation audit". Do NOT use for competitor business signals — use `competitor-intel` instead. Do NOT use for competitor messaging — use `competitor-positioning` instead. Do NOT use for general web scraping — use `nimble-web-expert` instead.
JLCPCB PCB fabrication and assembly — BOM/CPL generation, basic vs extended parts, assembly constraints, design rules, ordering workflow. Use with KiCad for JLCPCB manufacturing. Use this skill when the user mentions JLCPCB, wants to order PCBs or assembled boards, needs prototype bare PCBs and stencils, wants to know JLCPCB design rules and capabilities, or is asking about PCB manufacturing costs or turnaround times. For gerber/CPL export, stencil ordering, and BOM management, see the `bom` skill.
Watch for the 11 known AI-coding-agent failure modes (fabrication, scope_creep, security_vulnerability, etc.) — consult this skill before edits, dependency adds, completion claims, or anything that could trip a known supervision concern. Quote the snake_case failure-mode ids verbatim when flagging risks.
Turn messy user research notes, interviews, support tickets, surveys, and product context into an evidence-backed decision room: a single HTML artifact with an evidence ledger, theme map, confidence heatmap, opportunity matrix, decision memo, and experiment queue. Use when teams need to move from qualitative signals to product or design decisions without fabricating certainty.
Stop your AI from making things up. Use when your AI hallucinates, fabricates facts, isn't grounded in real data, doesn't cite sources, makes unsupported claims, or you need to verify AI responses against source material. Covers citation enforcement, faithfulness verification, grounding via retrieval, and confidence thresholds.