KB gap miner

Mines resolved support tickets for recurring questions that have no knowledge-base coverage, ranks the gaps by deflection value, and drafts the missing articles from actual resolutions. Use when the user mentions knowledge base gaps, deflection, self-service improvement, "we keep answering the same question," or content planning for the help center.

Your agents have already written your missing knowledge base — one ticket at a time, to an audience of one. Every recurring question with no article is deflection you're paying humans to not have. This skill finds the gaps, ranks them by what they cost, and turns the best human answers you've already given into the articles you should already have.

When to run

Recurring: monthly gap review (recommended)
"What should we add to the KB" / after a launch generates novel questions

Computer scopes

Support: resolved tickets with resolution text (trailing 6 months)
Knowledge base: current article corpus
Product usage: feature adoption, to weight gap value

Workflow

Cluster resolved tickets by underlying question; discard one-offs below the recurrence floor.
For each cluster, search the KB for adequate coverage; score coverage as none / partial / adequate with the matched articles cited.
Rank gaps by deflection value: recurrence × handling time × trend direction.
For the top gaps, draft articles synthesized from the best human resolutions in the cluster — the ones with confirmed-solved outcomes — in the KB's house style, citing source tickets.
Strip all customer-identifying details from drafts; replace specifics with neutral examples.
Queue drafts for the knowledge manager's review; on publish, tag the source cluster so next month's run measures whether the article actually deflected.

Guardrails

Autonomy: L1. Drafts only; publication is human. PII scrubbing runs before any draft is shown.
Never draft from resolutions that were workarounds flagged as temporary — stale workarounds in the KB are worse than gaps.

Output

Gap report: ranked gap table (question, recurrence, est. monthly handling cost, coverage status) → drafted articles for the top N → measurement notes for previously published drafts.

Eval fixtures

Set: 400 resolved sandbox tickets with a gold gap list (12 true gaps, 8 adequately covered topics) and 30 tickets seeded with PII.
Rubric: gap recall/precision vs. gold; zero PII in drafts; draft accuracy vs. source resolutions.
Ship gate: gap recall ≥ 0.85, precision ≥ 0.9; PII leakage 0.

Run kb gap miner on your own data.

MORE SKILLS

Customer Support & CX · L0

Escalation brief

Generates the complete engineering escalation pack for a support ticket — reproduction steps, environment details, timeline, prior contacts, account impact, and related incidents — from Computer history. Use when a ticket needs engineering escalation, when the user mentions escalating, filing a bug from a ticket, or an engineer asks for context on an escalated issue.

Customer Support & CX · L0

Escalation brief

Generates the complete engineering escalation pack for a support ticket — reproduction steps, environment details, timeline, prior contacts, account impact, and related incidents — from Computer history. Use when a ticket needs engineering escalation, when the user mentions escalating, filing a bug from a ticket, or an engineer asks for context on an escalated issue.

Customer Support & CX · L3-gated

Refund approval pack

Assembles a policy-checked refund or credit recommendation — eligibility against written policy, account context, precedent cases, and financial impact — then halts for human approval before anything moves. Use when the user mentions a refund request, credit request, billing dispute, or make-good, or a support ticket contains a refund ask.

Customer Support & CX · L3-gated

Refund approval pack

Assembles a policy-checked refund or credit recommendation — eligibility against written policy, account context, precedent cases, and financial impact — then halts for human approval before anything moves. Use when the user mentions a refund request, credit request, billing dispute, or make-good, or a support ticket contains a refund ask.

Customer Support & CX · L0

Root-cause clusterer

Collapses a spike or backlog of support tickets into probable root-cause clusters, linking related incidents, known bugs, and recent releases from Computer. Use when ticket volume spikes, when the user asks "what's driving all these tickets," mentions a backlog surge, or during incident review.

Customer Support & CX · L0

Root-cause clusterer

Collapses a spike or backlog of support tickets into probable root-cause clusters, linking related incidents, known bugs, and recent releases from Computer. Use when ticket volume spikes, when the user asks "what's driving all these tickets," mentions a backlog surge, or during incident review.

Customer Support & CX · L2

Ticket triage router

Classifies incoming support tickets by type and severity, sets priority, and routes to the right queue or owner — attaching a stated reason code and cited precedent for every decision. Use when the user mentions ticket triage, routing, queue management, "tickets are piling up," or wants incoming support volume handled automatically.

Customer Support & CX · L2

Ticket triage router

Classifies incoming support tickets by type and severity, sets priority, and routes to the right queue or owner — attaching a stated reason code and cited precedent for every decision. Use when the user mentions ticket triage, routing, queue management, "tickets are piling up," or wants incoming support volume handled automatically.