Audit (JSM)
JSMBootstraps project context by scanning a codebase and writing or gap-filling tool-agnostic AGENTS.md files.
By agent · Claude Code
Skills in the open SKILL.md format that Claude Code reads. Each one has a plain-language summary and the command to install it.
$ npx skills add <author>/<skill>Replace the placeholder with the author and skill name from the skill's page. Start a new Claude Code session afterwards so the skill is picked up.
Bootstraps project context by scanning a codebase and writing or gap-filling tool-agnostic AGENTS.md files.
Writes PR descriptions, changelog entries, release notes or postmortems from a project's commits, diffs and incident facts.
Scaffolds apps, libraries and other code in Nx workspaces using generators that match the repository's conventions.
Audits an active sales pipeline for stale deals, repeated close-date pushes and missing forecast-critical fields, then proposes deal-level dispositions and remediation.
Builds a RevOps or Sales Ops career plan based on role scope, competency gaps, verifiable impact evidence, interview readiness and compensation.
Audits sales pipeline stages against buyer-verifiable milestones and identifies rep-activity criteria, weak evidence and reporting risks.
Designs a RevOps revenue funnel model with stages, cohort conversion assumptions, ownership handoffs and planning inputs.
Designs a reconciling revenue KPI tree for B2B and B2C businesses, with owners, guardrails, review cadences and stage-specific metrics.
Diagnoses unreliable sales forecasts by reconstructing misses, separating data, behaviour and demand causes, and ranking fixes by impact and effort.
Builds hiring packets for RevOps and Sales Ops roles, covering role design, interviews, work samples, scorecards and ramp plans.
Designs field-level governance for CRMs, assigning owners, systems of record, freshness rules and enforcement for trusted revenue data.
Defines revenue data authority across systems with source-of-record and source-of-truth assignments, identity rules and metric ownership.
Designs CRM-based sales-to-CS handoff processes with gated triggers, required context, timing SLAs and clear ownership.
Builds and validates leading churn indicators from subscription account activity, product usage and support data into a ranked signal register.
Designs and validates weighted customer health scores that flag churn risk and expansion readiness from product, engagement, support and commercial signals.
Designs and validates revenue-operations lead scoring models using fit, engagement, thresholds and backtesting to improve MQL and PQL quality.
Structures RevOps revenue reports around locked metrics, finance reconciliation, plan-versus-actual analysis and clear board narratives.
Designs approval chains for non-standard B2B SaaS deals, covering concessions, margin floors, exceptions and escalation.
Reviews RevOps and GTM tool stacks to identify overlap, time renewal actions and decide which tools to keep, consolidate, replace or cut.
Creates evidence-backed creative briefs for advertising concepts, covering angles, hook directions, production specs, testing intent and handoff.
Designs CRM lead assignment systems covering account matching, territory rules, fair distribution, fallback queues and SLA-based escalation.
Verifies ad conversion tracking before launch, tracing test events, checking deduplication, consent, values and statuses, then issuing a GO or NO-GO decision.
Designs pre-launch paid-media creative test plans with hypotheses, budgeted test cells, sample sizing and pre-registered kill or scale rules.
Selects and sizes customer seeds for Meta, Google, LinkedIn and TikTok lookalike audiences, checking match rates, privacy and fallback options.
Questions
A folder with a SKILL.md file: instructions, and sometimes scripts and reference files, that Claude Code loads when a task calls for them.
Each skill page has an install command. Run it in a terminal, then start a new Claude Code session so the skill is picked up.
An instructions file is loaded in every session and describes the project. A skill is loaded only when a task calls for it, so it can be long and specific without costing context on every turn.