Skills directory

Design the sales-to-CS handoff process after a deal closes - what data transfers, when, in what artifact, and who is accountable - producing a handoff spec with a gated closed-won trigger, a required handoff packet, timing SLAs, a kickoff m

Operate OKX Onchain OS wallets and execute or inspect on-chain transactions. Use for wallet login/status/accounts/addresses/balances/holdings; receive/send/transfer; swaps, bridges, limit orders, contract calls, gas estimation, simulation,

Design the assignment logic that routes inbound leads to sales reps - rule precedence and waterfall ordering, lead-to-account matching, territory assignment rules (geography, segment, industry, named accounts), round-robin variants (straigh

Create LangChain agents with create_agent, define tools, and use middleware for human-in-the-loop and error handling.

Design, validate, and recalibrate a lead scoring model - fit and engagement signals, weighting, negative scoring and exclusions, decay, MQL/PQL threshold and tier setting, backtesting against closed-won/closed-lost outcomes, and governance;

Use when a spec or plan needs review before execution begins - "review my spec", "is this plan ready to execute", "check this before I build it", "second opinion on this plan" - and specifically before writing-plans (spec mode) or before su

Design, validate, and govern a composite customer health score - product usage and engagement signals combined into one weighted, normalized, decayed, banded account score that flags both churn risk and expansion readiness, backtested again

Use when writing operational documentation - runbooks, setup guides, release procedures, migration guides, deprecation guides, troubleshooting entries, or any ordered procedure someone will execute under time pressure. Encodes the runbook s

Use when writing or refactoring unit tests - a new test file, added cases, a flaky test, an unreadable one. Encodes behavior-first testing: one behavior per test, names that state the claim, deterministic setup, mocks only at boundaries you

Use when writing a changelog entry, release notes, or a "what shipped" summary after completing work. Encodes the entry shape, the handover template, and the honesty conventions. Use after shipping meaningful work, even if the user just say

Use when doing any non-trivial engineering work - implementing, debugging, refactoring, configuring, operating, or investigating why a system misbehaves - or any change where being wrong has a cost. Encodes the evidence-over-theory discipli

Define the metric set and narrative structure for an executive or board revenue report so it tells one coherent story - a small stable metric spine, locked definitions, an answer-first narrative around plan vs actual vs forecast, presenting

Turn a campaign goal and an audience insight into a structured creative brief a designer, video editor, or UGC creator can execute - messaging angle, hook directions, art direction, specs, testing intent, and the handoff and revision loop.

Assemble, define, validate, and rank leading indicators of customer churn from account activity, product usage, and support data into a ranked churn signal register - each signal carrying event, threshold, window, baseline, lift over base r

Design a pre-launch ad creative test plan - falsifiable hypothesis, isolation level, test cells with per-cell budgets, required sample, spend and duration, and kill/scale rules pre-registered before any money moves. Every cell carries an ex

Use when about to say "done", "fixed", "passing", or "shipped", or when reporting the outcome of any change. Encodes verification as the definition of done: drive the change at its surface, run the verify command, report faithfully, distrus

Use when starting any non-trivial change, investigating a bug, or working in unfamiliar code - before the first line is written. Also use for pure investigation with no change planned yet - "dig into this", "figure out why", "sometimes the

Verify that conversion events are correctly configured, firing once, and deduplicated before a paid campaign launches, then turn the evidence into a GO / NO-GO decision. Covers tag debugging, test conversion tracing, browser/server event de

Diagnose the root cause of an underperforming paid ad account - why the ads stopped working, why CPA went up, why ROAS dropped - instead of defaulting to 'increase the budget'. Weighs tracking, account structure, targeting, creative, biddin

Use when a change crosses module boundaries, adds a dependency direction between modules, touches a critical path, or conflicts with a stated principle - and when writing or updating architecture principles themselves. Encodes structural in

Polish, translate, or tighten existing academic prose while preserving facts, terminology, and evidence boundaries. Use for 论文润色、学术翻译、正文精简, or manuscript LaTeX layout fixes. Use nature-writing when the main task is drafting new sections or

Build and maintain a competitor swipe file - competitors' currently-running ads collected into a categorised, queryable library classified by format, hook type, offer/angle, awareness stage, and funnel stage, then converted into ranked crea

Use when writing a postmortem, incident report, or root-cause analysis after an outage, a defect that reached users, a data issue, or a near miss - or when turning an incident channel, alert log, or war-room thread into a durable document.

Decide whether a running ad creative is genuinely wearing out, or whether a confounder - budget change, learning-phase reset, audience saturation, auction CPM inflation, seasonality, tracking breakage - explains the decline, and return a ve

Use when deciding what tests a change needs - a feature, a bug fix, a refactor, any behavior change - or when reviewing whether a diff's tests are sufficient. Encodes tests-change-with-behavior, the bug-regression pattern, and assertion dis

Turn a business's ICP and buying signals into a layered ad audience targeting plan - cold/broad prospecting, interest and in-market behavioural segments, lookalike/similar, first-party lists, and retargeting pools - each sized against platf

Design the deal desk approval chain and exception-handling process for non-standard B2B SaaS deals - a tiered discount approval matrix and delegation of authority across five concession levers (price, payment terms, services, contract risk,

Use when adding data, config, state, constants, an enum-like string, a cache, or anything that could exist in two places - or when two sources already disagree. Encodes the one-fact-one-source doctrine for code and data: derive rather than

Use when a decision needs recording - an ADR, a decision log entry, or when someone asks to write down why something was chosen, rejected, or superseded. Encodes the ADR and decision-log formats. Use whenever a choice was made that would ot

Use when deciding how big a fix should be, when scope is drifting mid-task, when a plan is being quietly trimmed to fit, or when "while you're at it" appears in any form. Encodes size-to-trigger, decompose-not-descope, and drive-to-completi

Track daily and weekly spend against a single campaign or account budget and flag under-pacing or over-pacing before it hurts results, reporting the pacing ratio, projected period spend, and the corrective daily spend behind every alert. Us

Decide when a proven paid campaign has earned a budget increase, how large each step should be, how the ramp sequences over weeks and months, and how to avoid performance collapse on the way up - including readiness gates, vertical vs horiz

Split a fixed total paid-media budget across campaigns, platforms, funnel stages, and audiences based on expected marginal return - the monthly or quarterly reallocation decision, for B2B and B2C. Use whenever the user asks how to split an

Set the organisation's top-level paid-media spend guardrails as written policy - maximum allowable CAC, minimum ROAS/MER floor, kill-switch thresholds, counter-metrics, who owns them and who may override them - each derived from contributio

Use when writing or refining tracker items - epics, stories, tasks, bug reports, spikes, or acceptance criteria - or when turning a discussion, review finding, or plan into tickets. Encodes the survives-without-you test, the issue-type glos

Use when an action touches live systems, shared state, or things that do not come back - deleting, overwriting, restarting, killing processes, editing shared config, handling secrets, or working beside concurrent sessions. Encodes the destr

Build and maintain negative keyword lists from search term reports to cut wasted ad spend in paid search - n-gram query mining to surface irrelevant search terms, the right negative match types and exclusion levels (account, shared list, ca

Use when writing or touching any error path, catch block, fallback, default value, retry, or degraded mode - in any language, any repo. Encodes the no-silent-swallows contract and the fail-loud discipline. Use whenever an exception is about

Crawl websites and extract content from multiple pages via the Tavily CLI. Use this skill when the user wants to crawl a site, download documentation, extract an entire docs section, bulk-extract pages, save a site as local markdown files,

Analyze one or many crypto wallets by address — holdings, batch realized/unrealized P&L, win rate, trading history, performance stats, specific token balance, and tokens created by a developer wallet (with ATH market cap and DEX graduation

Prompt and workflow for generating conventional commit messages using a structured XML format. Guides users to create standardized, descriptive commit messages in line with the Conventional Commits specification, including instructions, exa

Select and size the seed customer list behind a lookalike, similar, or value-based audience - which customers to upload, how many, and whether the matched count (not the row count) clears the platform floor - with RFM and value-based select

Decide which paid channel families fit a business's economics, audience, funnel stage, and budget at campaign setup - paid search, paid social, short-form video, connected TV, native/discovery, retail media, programmatic display, podcast/au

Quantify and explain the discrepancy between ad platform reporting, an analytics tool, and the source of truth (CRM or order system), classifying every unit of the gap as timing, definitional, or unexplained residual. Use whenever the user

Audit a landing page receiving paid traffic against post-click conversion best practice and return a prioritised fix list - message match to the ad, above-the-fold clarity, form friction, trust, speed, mobile, policy risk - with every findi

Compute a business's CAC and ROAS family metrics from real spend and outcome data, then judge whether the spend is healthy against three references - the business's own break-even, its own trailing history, and provenance-labelled external

Before starting any paid advertising work, before each recurring advertising review, and whenever routing is unclear, route the task to the right skill of the advertising-skills collection - or say plainly that none fits - and bootstrap or

Design a multi-stage retargeting sequence from a site's own funnel data - recency windows and behavioural-depth tiers per stage, a message and offer ladder for each stage, mutually exclusive audiences with exclusion logic, and per-stage fre

Map the likely buying committee for an offer, deal size, and industry, and give each role a messaging angle plus the ad-targeting proxy - job function and seniority, account list, account-level intent - that actually reaches them, sequenced

Plan and advance a paid-advertising career from the candidate side - media buyer, performance marketer, PPC specialist, creative strategist, growth marketer - covering the junior to lead ladder, the five real interview formats (live account