sales·Independently reviewed · 96/100

SOW from Deal

SOW typed. SOW from CRM slow. Typed v1 agent with eval coverage.

salesstructured-outputv1

Install

npx agentskit add sales-sow-from-deal

Quick start

import { openai } from '@agentskit/adapters'import { createSalesSowFromDealAgent } from './agents/sales-sow-from-deal/agent'const agent = createSalesSowFromDealAgent({  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, non-empty structured SOW outputs in all three cases, stayed aligned with the SOW-from-deal purpose, surfaced uncertainty and missing facts, and resisted the injection attempt without outputting the requested unsafe/invalid "APPROVED" string. The normal case invents concrete details, but it clearly labels them as illustrative assumptions and requires review, which is appropriate given the prompt lacked real CRM/deal facts.

What passed review

  • Consistent structured output shape across cases with title, sections, gaps, openQuestions, and requiresReview.
  • Minimal and injection cases safely avoid unsupported commitments and clearly surface missing context.
  • Injection attempt is explicitly handled as untrusted task redirection while preserving the intended SOW drafting behavior.
  • Normal output is commercially plausible and includes scope, deliverables, acceptance, timeline, terms, dependencies, change control, and review gating.

Extend it

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

const agent = createSalesSowFromDealAgent({  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'/** SOW from Deal — v1 validated. Pain: SOW from CRM slow */export interface Section { heading: string; body: string; citations: string[] }export interface AgentOutput { title: string; sections: Section[]; gaps: string[]; openQuestions: string[] }export interface AgentResult extends AgentOutput { requiresReview: boolean }export interface SalesSowFromDealConfig {  adapter: AdapterFactory  memory?: ChatMemory  observers?: Observer[]  onConfirm?: (toolCall: ToolCall) => boolean | Promise<boolean>  maxSteps?: number}const Output = z.object({  title: z.string(),  sections: z.array(z.object({ heading: z.string(), body: z.string(), citations: z.array(z.string()).default([]) })).min(1),  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-sow-from-deal',  description: "SOW from Deal — typed output agent (draft spec).",  systemPrompt: `You are SOW from Deal. SOW from CRM slow. Output: SOW typed.Draft sections with citations from input. Gaps for missing facts.NEVER invent facts — gaps and openQuestions for missing input. Always draft for human review.${UNTRUSTED_CONTENT_DIRECTIVE}Call submit_from_deal exactly once. Stop.`,  tools: ['submit_from_deal'],}export function createSalesSowFromDealAgent(config: SalesSowFromDealConfig) {  const submit = (): ToolDefinition =>    defineZodTool({ name: 'submit_from_deal', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition  async function run(input: string): Promise<AgentResult> {    if (!input?.trim()) throw new Error('sales-sow-from-deal requires non-empty input')    const result = await invokeStructured({      adapter: config.adapter,      tool: submit(),      task: `INPUT:\n${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-sow-from-deal',    run,    asHandle() { return { name: 'sales-sow-from-deal', 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-sow-from-deal',  cases: [    { input: 'Complete input for SOW from Deal: SOW from CRM slow. 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