Frameworks and skills you can take away.

Two jobs, day to day. Refine: pull what is actually winning, find the patterns your current ICP misses or overstates, and propose tightenings you approve before anything is saved. Add a segment: define a new ICP segment alongside the existing ones when a genuinely distinct motion has emerged.

Builds or validates an ICP from evidence rather than opinion, using the GAP method and customer interviews. Treats the ICP as the highest-leverage thing to get right: when it is wrong, everything downstream inherits the error, from positioning and messaging to territory design, pipeline quality and CS segmentation.

Applies before any list gets built or bought. Produces a scored, tiered matrix that turns targeting from an opinion into a filter.

For when you are compared against a shelf of near-identical options and every conversation collapses to price. Look interchangeable and you get price-shopped. The output is a positioning package in three parts: a new way to frame the problem, category-defining language the market ties to you, and your expertise packaged as IP.

A six-section competitor battlecard in a consistent format: where you win, where you lose, the traps to set, and the talk tracks for each. Saved back to your knowledge base so the whole team works from one version rather than a rep's memory.

Treats every list build as a table: you define the columns, it fills them. Companies, contacts, or both, with qualification applied as it goes, so what comes out is ready for outreach rather than a raw dump you still have to clean.

A nine-step enrichment workflow with a provider waterfall (Apollo, Prospeo, LeadMagic, Findymail, then Clay's own patterns) so a miss at one provider falls through to the next instead of leaving a blank. Ships with 58 table templates.

Verify, enrich, and clean a contact or account list against Lusha's verified B2B database. Show every change with source and confidence level. Flag every record that can't be verified. Return a clean list and a data quality summary.

Pull a company's full verified profile from Lusha. Return firmographics, funding, tech stack, live buying signals, and contact coverage. One company in, one complete profile out.

Look up a contact's current verified profile via Lusha. Return their current title, verified email, direct dial, seniority, department, and tenure. Flag departures. Flag recent hires. Return one clean, verified record.

Full SEO topic research pipeline — from account understanding to a prioritized content roadmap ready for briefing. Runs seven phases in sequence: business context ingestion, opportunity extraction, keyword validation, SERP qualification, business fit filtering, topic shaping, and prioritization.

Take extracted opportunity angles and validate them against real search demand using Ahrefs. This is where Ahrefs comes in — not to dictate strategy, but to confirm that a business opportunity maps to actual search behavior.

Apply the commercial filter. This is the step that stops the keyword list from becoming a traffic play disguised as an SEO strategy.

Build a complete account intelligence document before any keyword research begins. The system needs to understand the account — what they sell, who they sell to, what they want to sell more of, and what already works — or it produces keyword output disconnected from commercial reality.

Use when AI-search visibility needs measuring: which engines mention the brand, for which buyer questions, citing which sources. Produces a visibility baseline, a trend line, and a ranked list of the sources worth influencing. Exact commands live in references/commands.md.

Use when citation data needs to become a content strategy. Produces a ranked list of the domains, pages, and content patterns that drive AI answers in the category — and where the brand can plausibly earn a mention. Work top-down; never stop at the domain level.

Use when a visibility score moves unexpectedly, a client wants receipts, or a scan result looks off. Produces an audit trail: aggregate → contributing scans → raw AI answers. Never explain a score movement from the aggregate alone — decompose first.

Turns your offer, your ICP and a why-now signal into cold email built on the R.E.P.L.Y. framework. Optimises for replies rather than opens, and leads with something useful instead of opening with the ask.

A complete operating system for cold email that gets replies instead of spam flags, tested across millions of sends. It assumes you already know copy is only one of three levers, and will not let you skip the other two.

A four-email campaign with a defined job for each send: a trigger-led opener, two follow-ups that add rather than repeat, and a break-up that gives the prospect an easy way to close the loop.

Thirty-four cold email templates, each carrying the reply rate it actually produced and the principle behind why it works, so you can pick on evidence rather than taste.

Five structures for a cold email, each with when to use it and how to time the send: what goes in the opener, how to earn the second line, and where the ask belongs.

Run this in the hour before a cold email campaign goes live. It produces a launch or hold decision, and the thresholds that will stop the campaign if the first days go wrong.

Run this before any new domain or mailbox sends cold email. It produces a ramp schedule and a clear go/no-go decision for opening campaign volume.

Reads every inbound reply, classifies the intent behind it (interested, not now, wrong person, unsubscribe, auto-responder) and drafts the response that fits, so a busy inbox does not quietly bury the replies that matter.

Give it a list of accounts. It reads every member of each account's sales team on LinkedIn and returns one brief per account: how they sell, how the team is shaped, how they position. A company with no public sales team is still a row, because that absence is itself a signal.

Produce a high-signal intelligence brief on a target account. Lead with a synthesized executive summary, suppress sections where data is thin, and tie next steps to specific people and signals surfaced during the session.

Turns raw signals (news, hiring, funding, tech stack, leadership moves, content) into a deep-dive account brief: who sits on the buying committee, what each of them cares about, and what to put in front of an executive before the meeting.

This is a Claude Code plugin by Sabahudin Murtic. Install it from his repo — https://github.com/sabahudin-web/competitive-intelligence-radar — which bundles the commands, agents, skills, and rules described here. It needs a BrightData API token (live web data) and a Notion connection (publishing), and the optional war room uses Claude Code's experimental agent-teams feature.

A fact-led brief on a named competitor, with every signal, news item and displacement angle shaped around the specific deal you are in. Built to be read before you walk into a competitive conversation, not filed afterwards.

A one-page executive brief on an account: who is involved, what has happened so far, what is outstanding, and what to raise next. Cuts routine noise and keeps the meetings and call notes that actually carry the story.