Bundles

All 9 Claude Skills Packs in one purchase — 55 installable skills across founder, ecommerce, trades, agency, studio operations, fractional exec, recruiting, support, real estate. 39% off the line.

Four tools for the whole retention lifecycle — onboarding sequences, support & success skills, churn autopsy kit, brand voice engine. One honest standard across every customer touch.

The whole retention lifecycle — onboarding, support & success, churn autopsy, brand voice — PLUS the AI Acceptable Use Policy Builder to govern the AI doing the work, with a 90-day playbook. The Customer Retention Bundle, plus governance.

Policy → People → Proof: the AI Acceptable Use Policy Builder, the AI Literacy & Workforce Training Kit, and the EU AI Act Readiness Kit — three governance tools run as one program with a 90-day playbook. A working aid, not legal advice.

Four tools for the whole selling motion — battlecards, fractional-exec skills pack, brand voice engine, meeting intelligence. Honest selling, on-brand, call to close.

The three Codex packs in one suite: Migration & Refactor, Code-Review & PR-Hygiene, and the Documentation Engine. 18 skills, 3 runnable CI gates. Fix it, guard it, document it.

The four-tool LLM dev-tools stack: RAG Retrieval Grader, Prompt Regression Lab, Prompt Injection Red Team Kit, and Agent Reliability Harness — one CI-ready tool for each way AI breaks, plus a pipeline playbook. Deterministic, offline, gate-your-build.

The audit-and-fix loop for AI visibility: the AEO Citation Audit & Optimizer Kit (diagnose where you stand) + the GEO / AI Visibility Playbook (fix it), with a 90-day loop playbook. Measure it, improve it, prove it. No guaranteed citations.

The honest back office: the AI Support Deflection Kit (escalate), the AI Document Extraction Kit (flag), and the Finance & Reporting Automation Kit (reconcile), with a connective playbook. One rule across inquiries, paperwork, and the books — don't guess, surface it. Humans stay in the loop.

The four ways a data file lies, each caught: schema & quality (CSV gate), cross-source agreement (reconciliation engine), formula integrity (spreadsheet auditor), and PII safety (structured-data scanner). Grade the file before you trust the number.

Prove how your AI is governed across four kinds of evidence: connector access (auditor), output records (audit-trail kit), disclosure (synthetic-content readiness), and data flows (vendor register). Answer with documents, not adjectives. A working aid, not legal advice.

Get AI spend and reliability under control in one pass: forecast the budget blowout, reclaim the unused seats inside the tools you keep, and catch the multi-step agents quietly failing end-to-end. Cheaper and reliable, not cheaper and broken.

Verify what an application claims — the document, never the candidate: a whole-application fabrication-risk triage plus a per-claim verification checklist. Triage which applications need checking, then verify them claim by claim. Every verdict routes to verification, never a reject. A working aid, not legal advice.

Keep production AI agents reliable end-to-end across the whole loop: evaluate the release at the trajectory level (reliability harness), catch the compounding quiet failures and drift in production (drift monitor), and close every incident for real (postmortem gate that won't close on an open critical fix). Evaluate, monitor, learn — the operations sibling to the dev-CI Reliability Bundle.

Build support deflection the honest way, then prove it's real. The AI Support Deflection Kit stands up deflection that always offers a person; the Support Deflection Honesty Grader audits whether your 'deflected' bucket is genuine resolution or hidden demand (abandonment + repeat contact) — REAL / SOFT / FALSE DEFLECTION, with a gate that forces FALSE when a third or more is hollow. Build it, then prove it — a deflection number you can defend.

Diagnose the cost of AI 'workslop,' fix the handoff that causes it, and prove the drop. Pairs the Workslop Cost & Verification-Tax Calculator (prices the hours a team spends verifying and redoing hollow AI output) with the Internal AI Handoff QA Gate (catches that work before it moves), plus a 90-day playbook that runs them together — measure, fix, prove. Two tools for less than buying each on its own.

Four gates for the four attack surfaces of an AI agent: what you install (MCP/skill trust), what it reads (prompt injection), what it remembers (context poisoning), and what it can reach (non-human identity). Secure the agent before you ship it.

Two sides of chargeback defense in one loop: the Chargeback Threshold & MID-Survival Gate ($89) computes your Visa VAMP and Mastercard ECP ratios against the thresholds you confirm and reads each account HEADROOM / APPROACHING / OVER THE LINE with a separate prevention scorecard; the Chargeback-Proof Order Evidence Kit ($49) matches representment evidence to the reason code the bank filed and verdicts each dispute SUBMIT-READY / STRENGTHEN / DON'T FIGHT with deadline and no-case overrides. A 90-day playbook connects them with the one rule neither tool makes alone — your ratio band sets your representment posture: over the line you fight everything winnable and pre-empt counted events, in headroom you're selective. Both full products plus the loop for less than buying separately. Grades numbers and evidence, never people. Not legal or payment-compliance advice.

Four RedHub systems that read the same plant from four angles — OEE honesty, PM execution, cost of poor quality, and critical-spares stockout risk — plus a 90-day playbook that runs them as one quarterly reliability rhythm. All four full products for less than buying separately. Every verdict from your own numbers; grades assets, never people.

A quote is a chain of four promises: worth chasing, priced right, lead time that holds, and buildable once they say yes. Four RedHub Systems check all four links — RFQ triage, cost calibration, backlog reality, order buildability — in the order a quote travels. $294 versus $346 separately, plus a connective playbook. Deterministic, offline.

An auditor doesn't check whether you have a quality system — they check whether it's telling the truth. Three RedHub Systems catch the three places it quietly isn't: NCR-to-CAPA routing, gauge calibration integrity, and ECN propagation. $209 versus $247 separately, plus a connective pre-audit playbook. Deterministic, offline.

Every relationship you depend on has a number attached, and both sides lie the same two ways. Two RedHub Systems split the deceptive single metric into honest reads: the Supplier Lead-Time Honesty Scorecard (inbound) and the Rep & Distributor Scorecard (outbound). $118 versus $138 separately, plus a connective playbook. Deterministic, offline, no dates.

A voice-agent guardrail can fail in three places: it was never there, it got edited out in the console, or a release quietly changed the behavior around it. Three RedHub Systems close all three — Voice Prompt Forge (written), Prompt Version Vault (live), Canary Call Library (behaving). $169 versus $217 separately, plus a connective playbook. Deterministic, offline.

Every call leaves a trail that carries three risks: you called someone you shouldn't have, you're holding a card number read aloud, or you're sitting on data you should have deleted. Three RedHub Systems cover the full life of call data — DNC Suppression Sweep, Spoken-PII Transcript Redactor, Retention Purge Scheduler. $179 versus $227 separately, plus a playbook. Deterministic, offline.

Every outbound call carries four compliance exposures that bite at different moments — wrong number, wrong time, kept PII, over-retained data. Four RedHub Systems close all four: DNC Suppression Sweep, Calling-Window Compliance Checker, Spoken-PII Transcript Redactor, Retention Purge Scheduler. $229 versus $296 separately, plus a connective playbook. Deterministic, offline.

The whole AI-voice-agent lifecycle under one honest standard: build, prove, and test the agent, run the live calls, and cover every call-data compliance exposure. All ten Voice Desk Systems, priced below sum-of-parts, plus the connective Voice Desk Playbook. $549 versus $800 separately. Deterministic, offline; every tool grades prompts, calls, and records — never a person.

Three ways to stop an AI agent, at the three moments it can do damage — Tool-Input Schema Turnstile (the door), Agent Action Admissibility Engine (the ledger), Validation-Failure Backstop (the catch) — plus a chain rollup that names the weakest layer. They fail independently; the bundle proves which link is weak.

One survey asks what you declared; two sweeps go and measure it. Domain Ontology Readiness Survey + Enum & Status-Value Drift Sweep + Entity Identity & Functional-Key Sweep, plus a claim-vs-evidence rollup returning CONFIRMED / KNOWN GAP / OVERSTATED where a claim and its measurement disagree.

The whole ontology / agent-safety line — all ten deterministic checks from whether your domain is written down to what the next change will break quietly, plus a lane rollup that names the worst finding and the next thing to run. A check you haven't run is not a check that passed.

Two halves of one door, plus the line's map. The Browser Line reads whether decisions are made on your server; Logged-In Is Not Allowed reads whether your server checks who is asking. Composed with AND, not worst-of — neither half alone is a door. DOOR HOLDS / ONE HALF OF A DOOR / NO DOOR.

Three kits that sit on top of each other — your tables' shape, the rules written on them, and what you're holding about people — read in dependency order. A clean rules verdict on a model that can't hold a rule is locked against nothing. DATA LAYER HOLDS / GAPS IN THE LAYER / NOTHING UNDERNEATH IT.

Three kits, and only one is about the money. Grades your order flow, then whether that grade described anything — a clean path proven against a sandbox on a shared account is PROVEN IN THE WRONG WORLD, ranked below a path that doesn't work. THE PATH IS PROVEN / NOT PROVEN / WRONG WORLD.

The whole Past the Prototype line — all ten production-readiness kits: the front door, the data layer, the money path, and whether you can get back and explain what you built. Everything to take a vibe-coded prototype to something you can charge for. $459 vs $600 separately.

All ten instruments in the distribution lane, plus the one thing none of them can do alone: a rollup that reads all ten answers together and names the binding constraint. Seven ordered stages from ground to spend; the earliest blocked stage is what is next, and every problem above it is marked NOT YET ACTIONABLE — real, but not yet worth touching. An instrument you did not run reads UNKNOWN and can never be the constraint.

All ten instruments of the execution-layer lane, plus a deterministic roll-up engine that reads their findings in order across three stages. A later stage can never read higher than an earlier one, so the roll-up names which of your readings are moot. Returns HELD END TO END, HELD IN PARTS or NOT HELD.