claude

quality-and-standards

Keep the work honest — canonical frameworks, verification, QA, judgment, closeout, shared memory, and the registry.

1 on GitHubMITupdated 4 days ago
/

Plugin quality-and-standards from marketplace local-service-spotlight in dennisyu/local-service-spotlight-skills. Commands read from the repo's manifest — 4 days ago.

What it installs

Skills (51)

  • agents-first-reassignment

    Use when reassigning work off an unavailable or non-delivering human — agents first, humans only if no authorized SOP/agent path; use the team's private assignment record and a verified agent-claim pattern.

  • ai-search-visibility

    Audit how AI answer systems describe a person or company, preserve dated answers and cited sources, identify unsupported or missing facts, and recommend source improvements. Use before a launch, deal conversation or reputation review; independently retest rather than promise rankings or inclusion.

  • boil-the-ocean

    Operating principles for the whole skill pack. Use when running any Local Service Spotlight skill: loop until the Definition of done passes, self-verify, compound with memory. Completeness is coverage, not model tier. Read before running any skill; this governs HOW they all execute. Pair with model-

  • business-brand-strategist

    Name what your company really is to its market — your ideal customers, service category, coverage area, and the one-sentence differentiation only this company can claim — backed by proof points from real jobs. Output is a positioning brief plus an authority scorecard baseline. Start every company-au

  • business-website-agent

    Build your company's entity home — your company name as the domain, a facts-first site with real service pages and consistent NAP (name, address, phone), structured so Google and AI treat it as the source of truth about the company. Use to create or audit a company website.

  • client-access-checklist

    Verify the scoped access and measurement prerequisites for client/site work: Search Console, the real edit path, analytics, tag management, DNS, business profiles, video, social and ads. Use at kickoff, before authorized publishing, or when checking whether an existing approved access route can be r

  • client-relationship-cadence

    Run recurring client-relationship maintenance without a scored metrics loop. Check for movement, perform one authorized safe increment, track asks and preserve state without unnecessary messages. Use when a relationship or mid-build project needs scoped continuity rather than a Personal Brand Score

  • content-agent

    The Content Agent - drop in one raw video and get back a lightly edited YouTube upload, a blog with schema, 5-10 short-clip picks, platform social posts, and an email - grounded in YOUR transcript and YOUR files. Honor the requested delivery mode; routine publication within recorded authority procee

  • content-factory

    Turn ONE recording into a full library — definitive article, clips, reels, posts, and ad-ready creative — through Produce, Process, Post, Promote, with every asset pointing at one canonical hub. Use when you have a recording to repurpose or a content system to design inside the Social Amplification

  • cross-post-watch

    Mirror the owner's original Facebook posts to X immediately and to LinkedIn as one best post per rolling 24 hours, word for word, with verified publication and a state file that prevents duplicates. Use when a personal-brand owner wants their own agent to distribute their Facebook posts to LinkedIn

  • definitive-article-writer

    Write or improve the canonical recipe for a repeatable task, or the canonical hub for a person, company, concept or product. Classify the page first, prove its claims, connect it to owned training and relevant assistance, and document the execution in a meta article.

  • dollar-a-day-strategist

    Amplify PROVEN organic content with Dennis Yu's Dollar-a-Day — $1/day for 7 days, kill the bottom 90%, then recommend $30 over 30 days behind winners. Use to turn posts with real engagement into reach, retargeting pools, and deals.

  • evidence-verification

    Establish what is actually TRUE about a client and prove it with records a stranger can check. Find the hard anchors hiding in public registries (Secretary of State, bar admission, ISBN, patents, ORCID, authority files), verify identity by matching a second fact rather than a matching name, and neve

  • fitting-reply

    Answer AI-written cold outreach in kind — a poem for a poem, wit for hype, plus the real qualifying questions — then archive the thread so your human never sees it. Switches off and hands over to a human the moment a real person shows up.

  • ga4-client-view

    Render a finished GA4 website MAA into the client-facing one-page executive summary (print-ready PDF + self-contained HTML), following the Local Service Spotlight display standard: stoplight coloring (red = critical/missing), a \"prepared by\" human owner, the client logo, the two standard charts (1

  • ga4-website-maa

    Produce the investigative Website / GA4 report for ANY small business using the Google Analytics MCP. A business-model router adapts the analysis to the client (local-service lead-gen, e-commerce, audience/list-building, membership/course, or donation/nonprofit) so it works for any vertical, never r

  • gct-screen

    Screen whether a business has clear Goals, Content, and Targeting plus independently checkable proof before it enters the Social Amplification Engine. Use for prospect qualification, SAE onboarding, or re-qualification after an offer, market, or capacity change.

  • geo-visibility-audit

    Measure how AI models actually see a person or brand — query ChatGPT, Gemini, Perplexity and Claude live, score identity vs discovery presence, name every citing source, and produce the weekly MAA GEO block. Use when asked for GEO rankings, AI citations, LLM visibility, AI share of voice, 'are we sh

  • google-ads-analyzer

    Turn a local service business's Google Ads data into a short weekly report (Metrics, Analysis, Action) that a project lead can act on, drafted under evidence rules and checked by a second agent before anyone sees it; trigger on "do the MAA for [client]", "weekly Google Ads MAA", "analyze my Google A

  • google-ads-change-scripts

    Generate one-time Google Ads Scripts that implement specific account changes from MAA recommendations, adding negatives, pausing ads/keywords, updating RSA copy, adjusting bids, or restructuring ad groups. Every script defaults to dry-run mode. Trigger on "implement these changes", "script to add th

  • google-ads-client-view

    Render a finished Google Ads MAA into a client-facing report. Produces a markdown client view AND a paste-ready HTML version (built for Basecamp, works in any rich-text editor), both led by a one-paragraph summary, two 13-week trend charts (leads, cost per lead), emoji-flagged critical items, and a

  • google-ads-copy-optimizer

    Write Google Ads RSA headlines, descriptions and asset text for a local service business where every line traces to a source: the client's own website (a dated fact sheet), the landing page the ad points to, the ad group's real search terms, or the results of copy that already ran. Fact-checks live

  • google-ads-daily-check

    Daily exceptions-only health check across a roster of Google Ads accounts: new disapprovals and policy or verification notices, campaigns that served yesterday and not today or carry a not-serving reason, budget caps and spend far outside the daily budget, conversion actions gone silent on campaigns

  • google-ads-lp-auditor

    Audit Google Ads landing pages for ad-to-page alignment, conversion friction, and Landing Page Experience signals. Use this skill whenever the google-ads-analyzer identifies Below Average Landing Page Experience on any keyword, when a campaign shows strong CTR but poor conversion rate (suggesting po

  • google-ads-onboarding

    Onboard a new Google Ads client into the weekly MAA system in one run: verify access, scrape the account's full history (12 to 24 months, all statuses, inherited state), investigate the website and landing pages, research the real geographic footprint, package the questions only the client can answe

  • google-ads-reviewer

    Harden a finished Google Ads MAA in a fresh context: re-apply the evidence gates to every finding and action, issue one verdict per item (KEEP, DEMOTE TO WATCH, DROP, REPULL) with the contradicting number quoted, then grade format against the MAA rubric and return a routed punch list. Trigger on "ha

  • google-ads-script

    Generate production-ready Google Ads Scripts (JavaScript) for automating campaign management, data pipelines, monitoring, and reporting. Trigger whenever someone asks to write, build, create, or fix a Google Ads script, including BigQuery sync, Quality Score tracking, search term mining, conversion

  • grokipedia-authority

    Assess evidence readiness for a person, company, podcast or book; prepare a source-backed Grokipedia submission package; and inspect existing pages for errors. Default to HOLD for submissions until the caller verifies the current supported route, action authority and any applicable operational hold.

  • knowledge-panel-entity-seo

    Audit entity identity, Person schema and corroboration; separately verify Knowledge Graph objects, visible Google panels and owner claim status. Produce evidence-backed fixes and an authorized-owner claim checklist without promising a panel or a review time.

  • measurement-analytics

    Collect and validate calls, leads, bookings, revenue, rankings, traffic, and ad spend for a weekly Metrics-Analysis-Action report. Use when connecting a local-service client's data sources, deciding API versus export, automating MAA across many clients, defining what a lead or call means, reconcilin

  • model-judgment

    The operating layer that makes boil-the-ocean affordable — let the agent use its OWN judgment to route each task to the cheapest model tier that clears the quality bar, delegating bulk work DOWN to smaller models in subagents and escalating hard judgment UP to the top model. Read alongside boil-the-

  • nine-triangles

    Apply the versioned Nine Triangles operational reference and resolve conflicting or reconstructed triangle names. Use when someone asks for the Nine Triangles, ACC versus AEC, SBP versus ABP, MAA, GCT, DDD, CID, LDT, CCS, or MOF; when mapping a marketing or operations problem to the framework; or wh

  • one-session-client-onboarding

    Use when a new client signs up or an onboarding is dragging across multiple calls. One session: agent prefills GCT, client confirms and grants access, first Friday MAA is the certainty date. Humans are not the routing layer.

  • outbound-action-closeout

    Verify and close out agent-authored email, Basecamp, DM, support, publishing, and other human-visible actions. Use when an agent drafts, sends, posts, publishes, writes a routine that may do so, or fixes a misrouted, unattributed, or unverified action.

  • personal-brand-audit

    Run a proof-backed inventory of a founder, executive, creator, or public figure and turn it into an exact 20-page visual audit plus an installable agent action plan. Use for personal-brand audits, credibility inventories, top-connection maps, reputation and AI/search visibility analysis, or a 30/60/

  • personal-brand-strategist

    Name who you really are to the market — your buy box, your passion, and the one-sentence differentiation only you can claim — by mining your own AI history, bio, and deal record. Start every personal-brand build here.

  • personal-brand-website-agent

    Build your entity home — your name as the domain, a facts-first homepage structured to earn a Knowledge Panel and become the one URL Google and AI treat as the truth about you. Use to create or audit a personal brand site.

  • positive-mentions-harvester

    Find and check praise for a person or firm. Use when reviews, press, talks or kind words are spread across sources. Keep one proof list and choose what is safe to share.

  • recursive-self-improvement-qa

    After every task execution, write a meta article with run evidence, compare it with the canonical recipe and prior runs, and propose supported improvements. Publish or change the recipe under existing authorization.

  • reputation-gap-analyzer

    Compare the reputation you HAVE (scored proof) against the reputation you NEED for your buy box, then run Metrics → Analysis → Action to ship a dated 30-day plan of content, proof, and boosts that closes the gap. Run after harvesting proof.

  • reuse-agent-access

    Reduce repeated sign-ins when setting up agents or moving work between AI apps. Discover and verify existing authorized connectors and browser sessions, use supported import or secure login only when needed, and publish reusable instructions without exposing credentials.

  • sales-every-day

    The Sales Every Day Agent — the capstone. Runs on a schedule (that's the point), reads YOUR funnel and YOUR weekly MAA, and comes back each morning with ONE staged selling action - the email drafted, the post written, the follow-up list ready. Honor requested draft delivery; complete routine sending

  • second-ring-network-mapper

    Maps an owner-authorized LinkedIn Connections.csv or export ZIP into a local, evidence-labeled relationship report without sending the file to Second Ring or Local Service Spotlight. Use when someone says "map my network," "run a Second Ring scan," "analyze my LinkedIn connections," "find a warm int

  • security-audit

    Continuously verify that a website is still the site you published — no injected spam, no rogue admin, no hidden plugin, no different page served to Googlebot than to humans. Use to stand up monitoring on a new property, to run a point-in-time compromise check, to investigate a "my site looks hacked

  • seo-audit

    Run a technical, content and authority SEO audit using the versioned seven-component Local Service Spotlight SEO & Growth rubric, with source-linked findings and UNKNOWN coverage. Use when a client asks "how is my SEO", before rebuilding a site, for monthly re-audits, or before promising a ranking o

  • shared-memory

    Read and update an authorized team's shared records, preserving one authority per record type and exact-revision handoffs. Use when an agent needs durable context, decisions, a fresh-session continuation, or a write-back that another agent can verify.

  • skill-registry

    Keep every Local Service Spotlight skill centrally available and activatable. Use when adding a new skill, auditing the skill inventory, wiring a skill into a scheduled job or agent, reconciling what is installed against the canonical pack, or when someone asks 'is this skill part of our framework',

  • social-amplification-engine

    Orchestrate Dennis Yu's six-phase Social Amplification Engine for an evidence-qualified, authorized client: Plumbing, Goals, Content, Targeting, Amplification, and weekly MAA optimization. Use after gct-screen and the engagement/roster gates, or to reconcile existing SAE components without creating

  • software-subscription-audit

    The Software Subscription Audit — a quarterly (scheduled) money-finding agent. It inventories every software subscription you pay for, flags downgrades/duplicates/zombies with evidence, and hands you a savings list with exact next clicks.

  • video-repurposing-agent

    Watch a YouTube channel, judge which new videos deserve repurposing, and turn winners into verified article packages on the right site — staged by default and published only under exact action authority. Use when a client or member wants real recordings feeding their Content Factory.

  • weekly-brand-maa

    Run a recurring Metrics → Analysis → Action report from a caller's verified brief, sources and delivery policy. Use for personal-brand or agency-roster metrics tracking, safe scoped updates, independent verification and a durable next-run record.

View source on GitHub localservicespotlight.com

More from dennisyu/local-service-spotlight-skills

All 5 plugins →

Personal-brand audit intake plus authority, reputation, and honest relationship-path diagnosis. Install lss-everything for the full workflow. (10 skills)

51 skills

1

Run the account — cadence, access, reporting including GA4 and Google Ads, audits. (23 skills)

51 skills

1

Turn one idea into everything — articles, video, repurposing, amplification, distribution. (9 skills)

51 skills

1

Everything we use, in one install — all 51 Local Service Spotlight skills. Start here; this is what most people want.

51 skills

1

Maintain quality-and-standards?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex plugins5 plugins