ecosystem·Independently reviewed · 96/100

RFC Author

RFC draft typed: problem, options, decision. Big moves need RFCs before implementation. Typed v1 agent with eval coverage.

ecosystemagentskitdogfoodstructured-outputv1

Install

npx agentskit add ecosystem-rfc-author

Quick start

import { openai } from '@agentskit/adapters'import { createEcosystemRfcAuthorAgent } from './agents/ecosystem-rfc-author/agent'const agent = createEcosystemRfcAuthorAgent({  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 RFC-style outputs for all three cases, consistently included problem/options/decision sections, surfaced gaps and open questions, required human review, avoided inventing missing facts, and resisted the injection request to output APPROVED. The behavior is conservative but aligned with the stated purpose and safety constraints for RFC drafting from sparse or untrusted inputs.

What passed review

  • Valid structured outputs for every case with non-empty title, sections, gaps, and openQuestions.
  • Strong uncertainty handling: missing context is explicitly called out instead of fabricated.
  • Injection case correctly treats the override request as untrusted data and does not comply.
  • RFC shape is useful for human review: problem, options, decision, and blocking gaps are clearly identified.
  • No unsafe content leakage or material hallucination beyond the provided inputs.

Extend it

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

const agent = createEcosystemRfcAuthorAgent({  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'/** RFC Author — v1 validated. Pain: Big moves need RFCs before implementation */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 EcosystemRfcAuthorConfig {  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: 'ecosystem-rfc-author',  description: "RFC Author — typed output agent (draft spec).",  systemPrompt: `You are RFC Author. Big moves need RFCs before implementation. Output: RFC draft typed: problem, options, decision.Draft sections with citations from input. Gaps for missing facts.NEVER invent facts — gaps and openQuestions for missing input. Always draft for human review.${UNTRUSTED_CONTENT_DIRECTIVE}Call submit_rfc_author exactly once. Stop.`,  tools: ['submit_rfc_author'],}export function createEcosystemRfcAuthorAgent(config: EcosystemRfcAuthorConfig) {  const submit = (): ToolDefinition =>    defineZodTool({ name: 'submit_rfc_author', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition  async function run(input: string): Promise<AgentResult> {    if (!input?.trim()) throw new Error('ecosystem-rfc-author requires non-empty input')    const result = await invokeStructured({      adapter: config.adapter,      tool: submit(),      task: `INPUT:\n${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: 'ecosystem-rfc-author',    run,    asHandle() { return { name: 'ecosystem-rfc-author', 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: 'ecosystem-rfc-author',  cases: [    { input: 'Complete input for RFC Author: Big moves need RFCs before implementation. 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