devangk003/gtm-agent-skills

32 skills

• ab-testing-messaging

  Design and run A/B tests on messaging variables with proper statistical discipline — Bayesian or frequentist depending on volume, minimum sample sizes, and one-variable-at-a-time design. Use when the user says "A/B test subject lines", "test opener variants", "which variant wins", or "sample size for A/B test.

  Needs a keyActs undeclared458 words
• campaign-management

  Monitor active multi-channel cadences in real-time — pull reply rate, bounce rate, complaint rate, sender reputation — and make pause/slow-down/swap-copy decisions against pre-defined thresholds. Surfaces A/B test results when statistically meaningful and auto-pauses on hard-threshold breach. Use when campaigns need ongoing oversight, metrics need daily/weekly review, A/B tests need decision-making, or reputation drift requires a pause.

  Needs a keyInstructions only563 words
• channel-performance

  Analyze per-channel performance and recommend budget reallocation using cost-per-meeting, cost-per-deal, marginal-CAC analysis, and Bullseye Framework refresh. Use when the user says "channel performance review", "which channel should get more budget", "compare channels", or "Bullseye refresh.

  Needs a keyActs undeclared427 words
• channel-strategy

  Identify and prioritize go-to-market channels using Weinberg's Bullseye Framework, CAC/LTV-per-channel math, channel-fit-by-ICP scoring, and stage-appropriate selection. Produces a focused 1–3 channel bet with experiment design and kill criteria. Use when the user says "which channels should we use", "we're doing too many channels", or "should we try TikTok for B2B.

  Needs a keyActs undeclared1,007 words
• cold-calling

  Generate cold-call scripts (9-second opener, discovery questions, gatekeeper-handler, ≤15-second voicemail) and orchestrate dial sessions with TCPA-compliant DNC scrubbing, quiet-hours enforcement, and call-disposition logging. Use when Tier-1 SAL-eligible prospects have verified phone_status, email+LinkedIn channels have plateaued, high-stakes accounts need live-voice contact, or multi-channel cadence requires a call leg.

  Needs a keyInstructions only579 words
• cold-email-sequence

  Write cold email sequences (5–7 touch, 14–21 day) using the CCQ copy framework, Pain-Trigger-Outcome openers, mobile-first formatting, and a hard ban on cliché openers. Gates every touch on the personalization-hook contract and email_status. Use when a Tier-1/2 lead list is enriched and scored, an ICP-grounded outbound burst is being planned, or an existing sequence's reply rate is below the 3% floor.

  Needs a keyActs undeclared1,239 words
• competitive-intelligence

  Operationalize competitor monitoring with signal scoring (Strength × Decision-relevance), tiered watch-lists, alert routing, intel-to-action workflow, and battle-card refresh cycles. Produces a continuous monitoring system. Use when the user says "track our competitors going forward", "set up competitor alerts", or "how often should we update battle cards.

  Needs a keyActs undeclared1,084 words
• competitor-analysis

  Map a competitive landscape by tiering competitors into Direct/Indirect/Substitute/Aspirational, profiling positioning/pricing/strengths/weaknesses, applying Helmer's 7 Powers and segment ownership, and producing 1-page battle cards. Use when the user says "who are our competitors", "build a battle card", or "what's our moat.

  Needs a keyActs undeclared1,090 words
• conversation-intelligence

  Extract structured intel from call transcripts and reply corpora — competitor mentions, pricing pushback, feature requests, champion language, blocker signals — and aggregate into pattern-frequency alerts. Use when the user says "extract intel from transcript", "pattern aggregation", "competitor mentions trending", or "feature requests this month.

  Needs a keyActs undeclared446 words
• crm-hygiene

  Audit and maintain CRM data quality — required-field gates, dedup, normalization, stale-record flagging, and orphan cleanup. Use when the user says "run hygiene audit", "dedup CRM", "normalize industries", or "pipeline gate failed on missing fields.

  Needs a keyActs undeclared403 words
• customer-feedback-analysis

  Extract themes from customer feedback (JTBD interviews, churn surveys, G2 reviews, support tickets) with verbatim discipline and route per theme class. Use when the user says "analyze customer interviews", "churn survey themes", "extract JTBD themes", or "product feedback patterns.

  Needs a keyActs undeclared505 words
• data-enrichment

  Enrich raw Lead records with verified emails, phone numbers, social links, and citable personalization hooks via a verifier waterfall, deliverability classification, and source-tracked hook capture. Use when a sourcing skill has produced a list with unverified emails, missing phones, or no hooks; when a BYO list needs schema repair; or when stale records need re-verification before outreach.

  Needs a keyActs undeclared1,019 words
• discovery-call-prep

  Produce a 1-page founder/AE briefing for an upcoming discovery call with MEDDPICC slots, discovery questions, objection prep, and competitive context. Use when the user says "prep me for this call", "brief me for the meeting", "discovery call tomorrow", or "refresh the briefing.

  Needs a keyActs undeclared425 words
• email-infrastructure-setup

  Set up and verify email infrastructure — dedicated outbound domains, SPF+DKIM+DMARC, RFC 8058 one-click unsubscribe, 14+ day warmup, and emit the readiness flag gating cold-email-sequence. Use when a new outbound program starts, deliverability has degraded, migrating sending platforms, or Google/Microsoft Feb 2024 enforcement breaks an existing sender.

  Needs a keyInstructions only943 words
• follow-up-management

  Schedule resume touches, nurture cadences, and no-show rescue flows for not-now, out-of-office, and warm-but-not-now recipients. Use when the user says "schedule follow-up", "parse resume date", "no-show rescue", or "nurture this lead.

  Needs a keyActs undeclared478 words
• handoff-protocol

  Hand off SAL-eligible leads from SDR to AE with a complete briefing package and SiriusDecisions SAL acceptance tracking. Use when the user says "hand off to AE", "handoff this deal", "SAL acceptance", or "SDR to AE transfer.

  Needs a keyActs undeclared446 words
• icp-definition

  Define a tiered Ideal Customer Profile using a 100-point weighted scorecard, Buyer/Champion/User/Blocker role mapping, Pain-Trigger-Outcome chains with workaround analysis, an anti-ICP boundary, and a trigger event library. Produces an ICP an SDR can apply in 2 minutes. Use when the user says "define our ICP", "who should we target", or "when should we disqualify.

  Needs a keyActs undeclared1,101 words
• icp-refinement-loop

  Refine the ICP scorecard from actual won/lost patterns after ≥30 closed deals — recompute tier cutoffs, re-tune dimension weights, surface segment shifts, propose ICP delta with evidence. Use when the user says "refine ICP", "recalibrate scorecard", "win rate doesn't match predictions", or "quarterly ICP audit.

  Needs a keyActs undeclared521 words
• kpi-reporting

  Produce a weekly GTM KPI report with north-star metric, leading + lagging indicators, what's-working/what's-not, WoW deltas, and recommended actions. Use when the user says "weekly GTM report", "board snapshot", "SDR metrics", or "what's working this week.

  Needs a keyActs undeclared468 words
• lead-scoring

  Score each Lead record against the ICP scorecard plus BANT and CHAMP qualification frameworks plus a trigger-strength-and-recency formula, write score + priority + tier + rationale onto the person record, and emit a per-run scoring interaction. Use when sourced + enriched leads need outreach prioritization, when a discovery-call decision needs evidence-backed qualification, or when the SDR/AE hand-off requires a tier label per record.

  Needs a keyActs undeclared1,205 words
• lead-sourcing-apollo

  Source B2B leads from Apollo by translating an ICP scorecard into firmographic and persona filters, quoting cost before pulling, normalizing to the canonical Lead schema, and pushing companies + persons + run-history interactions to the CRM. Use when the user wants to build an outbound list grounded in a defined ICP, has an Apollo seat or API key, has an Apollo CSV export, or has any list to ingest and normalize.

  Needs a keyActs undeclared1,193 words
• lead-sourcing-clay

  Source leads via Clay's table-based orchestration of Apollo, Hunter, Clearbit, and other data providers in a single workflow, with cost-aware multi-source chaining, dedup, and provenance tagging. Use when the user has a Clay seat or API key and wants multi-provider results in one pull, when single-source skills miss fields Clay can fill in-flow, or when a Clay table already exists and needs to push to CRM.

  Needs a keyActs undeclared1,113 words
• lead-sourcing-linkedin

  Source leads via LinkedIn Sales Navigator by translating role-based and trigger-based ICP criteria into Sales Nav search URLs and filter recipes, executing via session-based tools or manual CSV export, and normalizing to the canonical Lead schema with strict ToS compliance. Use when the play is role/trigger-driven, when Sales Nav surfaces signals Apollo can't see, or when LinkedIn URL is required as the dedup key.

  Needs a keyActs undeclared1,306 words
• lead-sourcing-web

  Source leads from the open web — job boards, news, press releases, RFPs, podcast guest lists, regulatory filings — by translating ICP triggers into web search queries and scraper jobs, normalizing results to the canonical Lead schema with citation-grade provenance. Use when the buying trigger is text-based, when target accounts are stealth-stage or regional, or when database-only sourcing produces empty firmographics.

  Needs a keyActs undeclared1,173 words
• linkedin-outreach

  Execute LinkedIn outreach via Sales Navigator + session-based tools — connection requests with ≤300 char notes, follow-up messages, InMails — with strict LinkedIn ToS compliance, weekly rate limits, and personalization-hook enforcement. Use when email is the wrong channel, LinkedIn URL is the strongest identity signal, the play is relationship-first social-selling, or multi-channel cadence requires a LinkedIn leg.

  Needs a keyInstructions only1,164 words
• market-research

  Research and size a target market by classifying market type, framing the category via JTBD, building a triangulated TAM/SAM/SOM with explicit confidence labels, scoring segments via a bowling-pin rubric, and identifying whitespace. Use when the user says "how big is this market", "help me size this", or "what category should we claim.

  Needs a keyActs undeclared961 words
• multi-channel-cadence

  Compose 5–9 touches across email+LinkedIn+cold call into a 14–21 day cadence — calling channel skills for copy/scripts, then orchestrating per-recipient sequencing with channel-isolation rules, capacity-cap respect, branch logic, and exit conditions. Use when Tier-1 needs more than email alone, channels are complementary, reply rate on single-channel has plateaued, or account-based plays require multi-thread coverage.

  Needs a keyInstructions only583 words
• objection-handling-library

  Match embedded objections to a 12-objection canonical library and produce ranked response variants with cadence-state recommendations. Use when the user says "handle this objection", "generate response variants", "match objection to library", or "we keep hearing X objection.

  Needs a keyActs undeclared436 words
• pipeline-stages

  Move deals through an 8-stage pipeline using deterministic stage-gate rules and MEDDPICC slot completion criteria. Use when the user says "advance this deal", "audit pipeline stages", "which deals are stuck", or "move to next stage.

  Needs a keyActs undeclared499 words
• positioning-strategy

  Define product positioning using Dunford's 5-component framework, JTBD-based wedge construction, message house architecture, value-by-role variations, and for-and-against wedges against named competitors. Use when the user says "help me position", "our positioning isn't working", or "how do we position against X.

  Needs a keyActs undeclared1,052 words
• reply-classification

  Classify inbound replies into a 9-label taxonomy with confidence scores and route to downstream skills. Use when the user says "triage replies", "classify inbox", "sort campaign responses", or "categorize inbound messages.

  Needs a keyActs undeclared476 words
• revenue-forecasting

  Forecast 30/60/90-day revenue using weighted-pipeline math, historical stage-conversion rates, and conservative/base/aggressive scenarios with explicit assumption sets. Use when the user says "forecast revenue", "board update numbers", "commit number", or "recalibrate conversion rates.

  Needs a keyActs undeclared424 words