Search: "actually"

Measures and interprets what a fuzzing campaign actually reaches, using llvm-cov, lcov, or a fuzzer's own coverage output. Covers baselining a new campaign, reading coverage reports, and turning uncovered regions into harness, seed, or dictionary work. Use when a fuzzer plateaus, when judging whether a harness is effective, after changing a harness, or when asking why some code is never reached.

A highly opinionated plugin marketplace for augmenting your Claude Code. Battle-tested skills and tools that actually work.

Use this skill for a one-off diagnostic across every available data operation for a brand, typically at account handover or kick-off: is the connection healthy, what can we actually read, is data flowing, and where are the gaps. It is the "audit the account before we start" pass, not a recurring report. Trigger on: "Audit Acme's programme", "Health-check the Acme account", "Run a full check on [brand]", "We just won Acme, what can we see?", "Is the Acme connection working end to end?".

Use this skill to measure the gap between partnerships on the books and partnerships producing anything, on either side of the market. For a publisher: how many programmes they have joined, how many actually earned in a period, and which are dead weight. For an advertiser or agency: how many publishers are on the programme's roster, how many produced a transaction, and how concentrated the revenue is among the few that did. The output is a read-only worklist; it never applies to, leaves, approves, or contacts anyone. Trigger on: "how many programmes have I joined?", "which of my affiliate programmes are dead?", "how many of our publishers actually sell anything?", "which publishers have never produced a sale?", "audit my joined programmes", "dormant programme sweep", "how concentrated is our partner revenue?".

Use this skill when a publisher or agency wants to actually apply to a batch of Awin brand programmes they have not yet joined. The joinable set is read from the Awin API; applying is a click Awin exposes no API for, so it is carried out in the operator's own authenticated Awin session through Claude-in-Chrome. For every programme the skill surfaces the advertiser's terms for review, lets the operator drop rows, and submits only after one explicit batch confirmation that stands as informed acceptance of the terms shown. Trigger on: "apply to these Awin brands", "join these Awin programmes for me", "submit my Awin applications", "apply to my Awin shortlist", "join [brand] on Awin".

Use this skill when an agency operator needs a presentation-ready Quarterly Business Review for a client brand: an executive-summary narrative, a multi-quarter trend (not just this quarter vs last), and a check of whether last quarter's recommended actions actually happened. This is the premium extension of the free `programme-performance-report` skill's QBR output profile, not a replacement for it. Trigger on: "Build the full QBR deck for Acme", "Prepare Acme's QBR with a trend and last quarter's action follow-up", "Get me presentation-ready for the Acme quarterly review", "Did we deliver on what we said last QBR for Acme?".

Best Cursor Rules is a curated collection of 33 high-quality Cursor AI rules that actually work. Includes a Setup Wizard that analyzes your project and generates customized rules automatically.