Agent Library · By function · Pipeline · Outbound · Proposals
Pipeline stops leaking at the follow-up.
Most mid-market sales teams don't lose deals to competitors. They lose them to the second email that never went out, the proposal that took nine days, and the CRM nobody updates. These agents carry the parts that depend on nothing but consistency.
Speed to lead
Inbound arrives across a form, an inbox and three DMs. By the time someone notices, the buyer has already spoken to someone faster.
The proposal tax
Every proposal is assembled from scratch by the person least able to spare the time — usually the founder or the CBO.
CRM as fiction
Reps update records when they remember. Forecasting runs on a dataset everyone privately knows is wrong.
The agents
6 agents. Each one earns its autonomy.
Agent
What it does
Autonomy
Human gate
ICP Prospector
Builds and scores target-account lists against your ICP, enriched with firmographics before a rep ever opens the record.
Auto
—
SDR Agent
Drafts personalised first-touch and follow-up sequences from real account context, not merge tags.
Notify
Human signs off before anything goes to a new logo
Proposal Architect
Assembles scoped proposals and pricing from a brief, using your templates and your past won deals.
Review
Pricing always signed off by a human
Lead Router
Classifies inbound by intent, fit and urgency, then routes to the right owner with the context already attached.
Auto
—
Meeting Brief
Pre-call briefs assembled from CRM history, email threads and public signals — in your inbox before the call.
Auto
—
Pipeline Hygienist
Flags stalled deals, missing next steps and records that contradict themselves. Drafts the nudge.
Notify
—
Autonomy key — Auto: executes above 95% confidence. Notify: executes at 85–95% with notification and undo. Review: always queued for a human. Flag only: never acts, alerts a person.
What changes
Three shifts, not a feature list.
Response time becomes a system property
Speed to lead stops depending on who happened to be at their desk. It becomes a number you set.
Proposals stop bottlenecking on one person
The assembly is automated; the pricing judgement stays human. Your most expensive person reviews instead of builds.
The forecast becomes trustworthy
Not because reps got more disciplined, but because hygiene stopped depending on discipline.
The eval bench
What these agents have to pass before they touch real work.
- Routing accuracy against a labelled set of 100 historic inbound leads
- Proposal drafts scored against your last 20 won deals for scope and pricing structure
- Zero fabricated facts about a prospect — every claim traceable to a source record
Written against your data, not ours. The bench is built during Blueprint and runs on every deploy from then on.
Related
Where this connects.
See it on your HubSpot.
A Blueprint maps which of these agents pays off first in your exact setup — and which to leave alone.
Apply for a Blueprint