Brand Logo

Agent-native prospecting · evidence first

Turn a market question into a prospect list you can interrogate

Give your AI agent an ICP, a territory, and the evidence standard. Clay returns structured research with source context and review states—without forcing another seat-based workspace into your stack.

Inspect a sample result

Copying is step one; you still run and configure the skill in your agent.

INSTALLnpx -y @okki-global/okki-go-taroball
SAMPLE TASK
Find B2B infrastructure teams hiring RevOps, then show the evidence behind each match.
AI agent prospecting research interface with evidence columns

Dated comparison

Compare the operating model, not a feature-count headline

Public claims change. This table records what can be checked, what depends on configuration, and what remains undisclosed as of 20 July 2026.

DimensionAgent-native skillConventional seat SaaSEvaluation question
Setup pathInstall inside an existing agentAccount and workspace onboardingWhere does daily work occur?
Task modelNatural-language, task scopedUI workflow and saved filtersCan the method be reproduced?
Human reviewExplicit approval checkpointsConfig-dependentWho approves targeting and sending?
Data disclosureSource context in resultVaries by providerCan a reviewer trace the record?
Cost modelRuntime and provider dependentUsually seat or credit basedWhich unit drives marginal cost?
Security detailInspect permissions before runningReview vendor documentationWhere are keys and outputs stored?

“Varies” and “config-dependent” are not negative scores; they indicate facts that require environment-specific verification.

Auditable preview

Every shortlist should carry its own review trail

SAMPLE

Prospect result

Northstar Systems

ICP match: mid-market infrastructure software. Signals: RevOps role posted; CRM migration mentioned.

Vector Ledger

ICP match: finance operations platform. Signals: territory expansion; sales operations hiring.

Research notes

Compact references for a defensible workflow

01

ICP evidence worksheet

Define firmographic, technographic, exclusion, and confidence fields before a search begins.

Read method →
02

Human-review checklist

Inspect relevance, source recency, suppression rules, and regional outreach obligations.

Review controls →
03

Operating-model comparison

Evaluate seat SaaS, waterfall enrichment, browser automation, and agent-native execution.

Open comparison →

Four-state runbook

Copy is not install; install is not a first result

  1. 01

    Copy

    npx -y @okki-global/okki-go-taroball

    Success signal: the exact command is in your clipboard.

  2. 02

    Run

    Paste the command into the terminal used by your supported AI-agent environment. Read the package prompt before allowing execution.

    Success signal: the installer completes without a package or permission error.

  3. 03

    Configure

    Add only the provider credentials your chosen workflow requires. Keep keys in the runtime secret store; never paste them into prospecting prompts.

    Success signal: the agent can enumerate the installed skill without exposing secrets.

  4. 04

    First result

    Begin with a narrow ICP and request evidence fields. Review sources, uncertain matches, and exclusions before exporting or drafting outreach.

    Success signal: a structured sample appears with reviewable provenance.

Install FAQ

Questions to settle before the first run

It invokes the published installer through npx. Review the terminal output, requested permissions, package provenance, and your runtime policy before continuing.

No. Copy only places the exact text on your clipboard. You decide when and where to run it.

The recommended workflow keeps targeting, message approval, suppression, and send decisions under human control. Follow CAN-SPAM, CASL, GDPR, and other rules applicable to your recipients and region.

Use your agent runtime’s secret management mechanism, scope permissions narrowly, rotate credentials, and confirm deletion behavior for any stored result.