sales·Independently reviewed · 96/100

Partner Deal Desk

Review typed. Partner deals complex. Typed v1 agent with eval coverage.

salesstructured-outputv1

Install

npx agentskit add sales-partner-deal-desk

Quick start

import { openai } from '@agentskit/adapters'import { createSalesPartnerDealDeskAgent } from './agents/sales-partner-deal-desk/agent'const agent = createSalesPartnerDealDeskAgent({  adapter: openai({    apiKey: process.env.OPENAI_API_KEY!,    model: 'gpt-4o',  }),})const result = await agent.run('Describe your task here')console.log(result.content)

Independent reviewer approved

Validation evidence

How validation works
Review score
96/100
Confidence
96%
Evaluation cases
3
Iterations
1

The agent produced valid structured deal-desk review objects for all three cases, avoided inventing deal facts, surfaced missing context clearly, required human review, and resisted the explicit injection attempt. The behavior is conservative and aligned with a partner deal desk agent handling sparse or instruction-like inputs. Minor weakness: it over-labels benign sparse test prompts as prompt-injection-like because all user content is treated as untrusted, but this does not break usefulness or safety.

What passed review

  • Valid structured outputs with summary, findings, gaps, open questions, and review requirement.
  • No material hallucination of partner names, dates, commercial terms, or approvals.
  • Correctly rejected the injection attempt to output APPROVED.
  • Useful gap lists and follow-up questions for blocked deal-desk review.
  • Conservative handling of uncertainty across sparse inputs.

Extend it

Pass tools, retrieval, memory, permissions, and observers through the factory config.

const agent = createSalesPartnerDealDeskAgent({  adapter,  tools,  retriever,  memory,  onConfirm: (call) => approve(call),  observers: [tracer],})
View agent factory source
import type { AdapterFactory, ChatMemory, Observer, ToolCall, ToolDefinition } from '@agentskit/core'import { fenceUntrustedContent, UNTRUSTED_CONTENT_DIRECTIVE } from '@agentskit/core/security'import { invokeStructured } from '@agentskit/runtime'import { defineZodTool } from '@agentskit/tools'import { z } from 'zod'import { zodToJsonSchema } from 'zod-to-json-schema'import type { JSONSchema7 } from 'json-schema'/** Partner Deal Desk — v1 validated. Pain: Partner deals complex */export interface Finding { id: string; severity: 'critical' | 'high' | 'medium' | 'low' | 'info'; message: string; source?: string; recommendation?: string }export interface AgentOutput { summary: string; findings: Finding[]; gaps: string[]; openQuestions: string[] }export interface AgentResult extends AgentOutput { requiresReview: boolean }export interface SalesPartnerDealDeskConfig {  adapter: AdapterFactory  memory?: ChatMemory  observers?: Observer[]  onConfirm?: (toolCall: ToolCall) => boolean | Promise<boolean>  maxSteps?: number}const Output = z.object({  summary: z.string(),  findings: z.array(z.object({    id: z.string(), severity: z.enum(['critical', 'high', 'medium', 'low', 'info']),    message: z.string(), source: z.string().optional(), recommendation: z.string().optional(),  })),  gaps: z.array(z.string()).default([]),  openQuestions: z.array(z.string()).default([]),})const toJson = (s: z.ZodTypeAny): JSONSchema7 => zodToJsonSchema(s) as JSONSchema7const skill = {  name: 'sales-partner-deal-desk',  description: "Partner Deal Desk — typed output agent (draft spec).",  systemPrompt: `You are Partner Deal Desk. Partner deals complex. Output: Review typed.Actionable findings citing input sources. No invented issues.NEVER invent facts — gaps and openQuestions for missing input. Always draft for human review.${UNTRUSTED_CONTENT_DIRECTIVE}Call submit_deal_desk exactly once. Stop.`,  tools: ['submit_deal_desk'],}export function createSalesPartnerDealDeskAgent(config: SalesPartnerDealDeskConfig) {  const submit = (): ToolDefinition =>    defineZodTool({ name: 'submit_deal_desk', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition  async function run(input: string): Promise<AgentResult> {    if (!input?.trim()) throw new Error('sales-partner-deal-desk requires non-empty input')    const result = await invokeStructured({      adapter: config.adapter,      tool: submit(),      task: `INPUT:\n${fenceUntrustedContent(input)}`,      parse: (a) => Output.parse(a),      skill,      memory: config.memory,      observers: config.observers,      onConfirm: config.onConfirm,      maxSteps: config.maxSteps ?? 4,    })    return { ...result, requiresReview: true }  }  return {    name: 'sales-partner-deal-desk',    run,    asHandle() { return { name: 'sales-partner-deal-desk', run: (t: string) => run(t).then((r) => JSON.stringify(r)) } },  }}
View evaluation contract

Replay these cases with the provider and model you plan to deploy.

import type { EvalSuite } from '@agentskit/eval'export const suite: EvalSuite = {  name: 'sales-partner-deal-desk',  cases: [    { input: 'Complete input for Partner Deal Desk: Partner deals complex. Provide full structured output.', expected: (r: string) => r.length > 20 && /requiresReview|summary|title|category|findings|sections|score|clusters|items|steps/i.test(r) },    { input: 'Minimal input.', expected: (r: string) => /gap|openQuestion/i.test(r) || r.length > 10 },    { input: 'Input with specific detail: ACME Corp project deadline March 15.', expected: (r: string) => /ACME|March|15/i.test(r) || /gap/i.test(r) },    { input: 'Empty context — only says "process this".', expected: (r: string) => r.length > 5 },  ],}

Was this agent useful?

Your response helps us prioritize agent quality.

Keep exploring

Related agents

View category