Chatchat Skills

Deploy applications and infrastructure to Cloudflare using Workers, Pages, and related platform services. Use when the user asks to deploy, host, publish, or set up a project on Cloudflare.

Use this skill when automating GitHub workflows for issues, pull requests, labels, releases, repository hygiene, or operational guardrails around GitHub actions and repository maintenance.

Use this skill when leading or structuring incident response, including triage, containment, communications, recovery, severity assessment, timelines, and post-incident follow-up.

Use this skill when organizing invoices and receipts for tax preparation by extracting key fields, normalizing filenames, grouping records, and producing accountant-ready CSV output.

Use this skill when checking token security before swaps or trades, including honeypot risk, scam indicators, malicious contract functions, dangerous ownership settings, unusual taxes, and supported-chain audit data.

Knowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."

Use this skill when configuring or debugging Turborepo local and remote caching, including task outputs, cache keys, CI reproducibility, cache misses, and developer build performance.

Use this skill when upgrading dependencies with risk-aware batching, compatibility checks, lockfile review, migration notes, and verification steps for routine updates or major-version changes.

Use this skill when providing respectful, non-explicit intimate wellbeing guidance focused on practical help, communication, stress, sleep, exercise, habits, medical red flags, relationship health, or sexual performance concerns.

Use this skill when creating structured meal plans, calorie targets, macro calculations, meal timing, or goal-specific nutrition plans.

Use this skill when querying on-chain wallet addresses for active token positions, balances, token metadata, prices, 24-hour change, holding quantities, and portfolio summaries across supported chains.

Use this skill when deploying to Railway, including environment checks, service configuration, health verification, rollout observation, logs, and rollback readiness.

Use this skill when designing load balancing or traffic distribution for AI agents, including routing policies, reliability targets, latency control, throughput scaling, and demand-aware failover.

Use this skill when designing, reviewing, or validating event triggers, schedule triggers, automation jobs, CI/CD pipeline triggers, deployment workflows, and their edge cases.

Use this skill when preparing GitHub pull requests with clear summaries, scoped changes, test plans, reviewer context, risk notes, and links to relevant issues or evidence.

Use this skill when designing or reviewing Infrastructure as Code for reliability, security, drift control, modularity, environment promotion, and operational maintainability.

Help address review/issue comments on the open GitHub PR for the current branch using gh CLI; verify gh auth first and prompt the user to authenticate if not logged in.

Use this skill when creating or maintaining reusable infrastructure, CI/CD, service, or pipeline templates with governance, versioning, rollout consistency, and environment provisioning.

Use this skill when designing or reviewing Apache Airflow DAGs for scheduling, retries, task dependencies, idempotency, backfills, sensors, and production reliability.

Use this skill when designing, modeling, or operating BAP-578 agent economies where NFAs transact, collaborate, compete, provide services, handle funding, build reputation, or align incentives.

Use this skill when generating or maintaining release changelogs from merged work, including version scope, grouped changes, breaking-change notes, formatting consistency, and verification checks.

Use this skill when integrating BAP-578 smart contract reads or writes into a Next.js or React frontend, including wallet connection, chain setup, ABI wiring, hooks, minting flows, and agent-management UI.

Use this skill when auditing, hardening, or verifying the security posture of BAP-578 Non-Fungible Agent deployments, including threat models, NFA-specific vulnerabilities, and remediation guidance.

The Budget Pacing Monitor is a real-time budget tracking and forecasting agent that monitors budget consumption, identifies pacing issues, and recommends or implements budget adjustments to ensure optimal delivery. This skill provides continuous surveillance of daily and lifetime budgets across campaigns and ad groups, calculating pacing variance, forecasting budget exhaustion dates, and alerting marketers to critical over-pacing or under-delivery scenarios. It goes beyond simple spend tracking by diagnosing root causes (low bids, narrow targeting, learning phase issues) and providing corrective actions with forecasted delivery impact.

Use this skill when troubleshooting DevOps failures across CI/CD, infrastructure, runtime services, and deployment pipelines with evidence-driven incident triage and remediation steps.

Create a new blank Figma file. Use when the user wants to create a new Figma design or FigJam file, or when you need a new file before calling use_figma. Handles plan resolution via whoami if needed. Usage — /figma-create-new-file [editorType] [fileName] (e.g. /figma-create-new-file figjam My Whiteboard)

Use this skill alongside figma-use when the task involves translating an application page, view, or multi-section layout into Figma. Triggers: 'write to Figma', 'create in Figma from code', 'push page to Figma', 'take this app/page and build it in Figma', 'create a screen', 'build a landing page in Figma', 'update the Figma screen to match code'. This is the preferred workflow skill whenever the user wants to build or update a full page, screen, or view in Figma from code or a description. Discovers design system components, variables, and styles via search_design_system, imports them, and assembles screens incrementally section-by-section using design system tokens instead of hardcoded values.

Repository-grounded threat modeling that enumerates trust boundaries, assets, attacker capabilities, abuse paths, and mitigations, and writes a concise Markdown threat model. Trigger only when the user explicitly asks to threat model a codebase or path, enumerate threats/abuse paths, or perform AppSec threat modeling. Do not trigger for general architecture summaries, code review, or non-security design work.

Use this skill when drafting or reviewing MySQL or PostgreSQL migration scripts, schema changes, data backfills, index changes, permission updates, or operational SQL where locking, transactions, compatibility, rollout sequencing, validation, and rollback risk matter.

Use this skill when designing, building, deploying, binding, or reviewing logic contracts that define autonomous on-chain behavior for BAP-578 Non-Fungible Agents.

Use this skill when crafting BAP-578 AgentMetadata identity data, including personas, experience fields, vault schemas, voice profiles, metadata URIs, vault integrity, and rich verifiable agent identities.

Build or update a professional-grade design system in Figma from a codebase. Use when the user wants to create variables/tokens, build component libraries, set up theming (light/dark modes), document foundations, or reconcile gaps between code and Figma. This skill teaches WHAT to build and in WHAT ORDER — it complements the `figma-use` skill which teaches HOW to call the Plugin API. Both skills should be loaded together.

Translates Figma designs into production-ready application code with 1:1 visual fidelity. Use when implementing UI code from Figma files, when user mentions "implement design", "generate code", "implement component", provides Figma URLs, or asks to build components matching Figma specs. For Figma canvas writes via `use_figma`, use `figma-use`.

Use this skill when designing observability for services with logs, metrics, traces, alert thresholds, dashboards, SLO signals, and actionable incident-response context.

Analyze git repositories to build a security ownership topology (people-to-file), compute bus factor and sensitive-code ownership, and export CSV/JSON for graph databases and visualization. Trigger only when the user explicitly wants a security-oriented ownership or bus-factor analysis grounded in git history (for example: orphaned sensitive code, security maintainers, CODEOWNERS reality checks for risk, sensitive hotspots, or ownership clusters). Do not trigger for general maintainer lists or non-security ownership questions.

Platform-neutral TikTok account benchmark workflow. Use whenever the user asks whether a Campaign, Ad Group, Ad/Creative, or account result is good, strong, weak, worth scaling, better/worse than their account, above/below account baseline, or needs same-account relative performance context — even if they do not say "benchmark". Triggers: account baseline, relative performance, is this good, small sample noise, which one is stronger, 同账户对比, 账户基准, 表现好不好, 还行吗, 值不值得继续投, 哪个更靠谱, 是不是小样本假象, 高 CVR 是否可信. This read-only skill fetches TikTok reports through tt-ads MCP and computes deterministic same-advertiser, same-objective, same-grain benchmark statistics locally. Do NOT use for raw report pulls only (route to tt4b-get-performance-report), delivery/root-cause diagnosis such as no spend or rejection (route to tt4b-diagnose-campaign-health), or budget reallocation/execution (route to tt4b-optimize-budget).

The Campaign Health Auditor is a comprehensive diagnostic agent that conducts multi-layered health checks across campaigns, identifying issues, opportunities, and providing prioritized recommendations. This skill performs end-to-end campaign analysis—from account structure and approval status to creative fatigue, delivery diagnostics, and performance benchmarking. It generates a quantified health score (0-100), categorizes findings into Critical Issues, Warnings, and Opportunities, and delivers actionable next steps prioritized by impact and effort. Perfect for regular campaign reviews, client reporting, or pre-flight checks before major launches.

Use this skill when designing or troubleshooting hybrid cloud networking across on-prem and cloud environments, including secure connectivity, routing, VPN or Direct Connect style links, and traffic diagnostics.

Use this skill when reasoning about individual PostgreSQL tables and nearby relationships, including keys, constraints, column types, normalization, query-aware indexes, and migration-ready schema decisions.

Perform language and framework specific security best-practice reviews and suggest improvements. Trigger only when the user explicitly requests security best practices guidance, a security review/report, or secure-by-default coding help. Trigger only for supported languages (python, javascript/typescript, go). Do not trigger for general code review, debugging, or non-security tasks.

For TikTok Shop advertisers, identify high-potential remarketing audiences from real reach, audience insight, and existing DMP audience metadata.

Use this skill when reviewing Azure SDK management-plane pull requests for API compatibility, naming conventions, generated-code expectations, breaking changes, and SDK quality standards.

Produces natural-language creative recommendations and 2-3 video directions for the Hot Trend Creative Agent. Use when the user asks what creative direction may work now, what trends to use, how to refresh an active campaign, or how to restart an inactive advertiser based on advertiser status, historical creative signals, hot trends, or possible creative fatigue.

Use this skill when adding mode-aware behavior to an AI agent, including explicit triggers, permissions, response contracts, and plan, ask, debug, or execute style interaction modes.

Use this skill when designing or evaluating multi-cloud architectures, workload placement, shared control planes, failover patterns, vendor-risk tradeoffs, compliance constraints, or cross-cloud resilience.

Capture conversations and decisions into structured Notion pages; use when turning chats/notes into wiki entries, how-tos, decisions, or FAQs with proper linking.

Use when a TikTok advertiser, media buyer, creative strategist, brand operator, or growth team wants an AI agent to turn a product, Amazon page, ecommerce URL, landing page, offer, or ad brief into TikTok-ready ad concepts, hooks, storyboards, voiceover copy, AI image/video prompts, editing briefs, compliance review, delivery checklists, or a confirmed finished ad production workflow.

Use this skill when asked about Airwallex cash position — balances, receivables, obligations, FX exposure, runway, indicative FX rates — or to correctly refuse a money-movement request.

Use this skill when analyzing AI agent performance with latency, success rate, cost, quality, throughput, and failure signals to identify bottlenecks and improvement opportunities.

The Conversion Funnel Analyst is a sophisticated funnel optimization agent that helps users understand their conversion funnel, identify where users drop off, and recommend optimizations to improve conversion rates. This skill maps multi-stage funnels from impression to conversion, calculates drop-off rates at each stage, benchmarks performance against industry standards, and segments analysis by device, placement, audience, and creative type. It goes beyond reporting by diagnosing root causes for leak points—whether due to ad-to-page mismatch, pricing friction, or checkout complexity—and forecasts the impact of proposed optimizations on overall conversion volume and CPA.

Use this skill when designing deployment pipelines with safe promotion stages, quality gates, environment consistency, rollout strategy, rollback paths, and release verification.

Use this skill when provisioning virtual or physical corporate cards in Airwallex Issuing — create cardholders, issue cards with spend limits, and manage card spending. Not for transfers or invoices.

Use this skill when building or reviewing browser extensions that need extension-specific guidance for runtime boundaries, permissions, content scripts, background scripts, messaging, storage, security, or distribution.

Use this skill when scanning Cairo or StarkNet smart contracts for StarkNet-specific vulnerabilities such as felt252 arithmetic overflow, L1-L2 messaging flaws, address conversion bugs, or signature replay.

Prepare meeting materials with Notion context and Codex research; use when gathering context, drafting agendas/pre-reads, and tailoring materials to attendees.

Use this skill when decomposing and coordinating multi-step implementation or DevOps work across focused subagents with explicit ownership, handoffs, validation, and final integration checks.

Use this skill when a user wants to evaluate the performance of an active TikTok campaign and get prioritized recommendations for how to optimize it. This is a mid-flight evaluation tool for enterprise activators — it pulls live campaign data, assesses performance against KPI goals and vertical benchmarks, and produces an ordered list of optimization actions the activator (or the TikTok MCP) can execute immediately. Trigger phrases include: "evaluate the campaign", "evaluate campaign performance", "optimize my campaign", "optimization recommendations", "why is my campaign underperforming", "how should I optimize", "what changes should I make", "mid-flight evaluation", "what's wrong with my campaign", "campaign health check", "TikTok optimization", "how do I improve performance", "campaign isn't hitting goal", "too expensive", "low CTR", "low delivery", "creative fatigue", "vertical benchmarks for TikTok", or any request combining campaign performance review with recommended actions.

Syncs leads from TikTok Instant Forms into HubSpot CRM contacts. Covers ad account discovery, time window selection, standard field mapping, and email-based deduplication. Also guides users toward the native HubSpot Ads integration for ongoing automated sync. Use when the user wants to sync TikTok leads to HubSpot, mentions TikTok lead gen forms, wants to get TikTok leads into their CRM, asks about a historical lead backfill, asks whether they have received any TikTok leads, asks what leads came from TikTok ads, or asks about the status of their TikTok lead gen forms.

Use this skill when coordinating multiple concurrent features, stacked PRs, branch strategy, dependency sequencing, integration checkpoints, feature flags, or release timing without destabilizing main.

Use this skill when creating Airwallex invoices or subscriptions from a purchase order, contract, or quote — extract line items and match existing customers, products, and prices.