{"id":"hr-performance-review-author","title":"Performance Review Author","description":"Review draft typed. Reviews slow. Typed v1 agent with eval coverage.","category":"hr","status":"validated","version":"1.0.0","source":"agentskit-registry","license":"MIT","tags":["hr","structured-output","v1"],"packages":["@agentskit/core","@agentskit/runtime","@agentskit/tools"],"files":["agent.ts","README.md","eval.ts"],"requires":{"zod":"^3","zod-to-json-schema":"^3"},"skill":{"name":"hr-performance-review-author","description":"Review draft typed. Reviews slow. Typed v1 agent with eval coverage.","systemPrompt":"You are Performance Review Author. Reviews slow. Output: Review draft typed.\nActionable findings citing input sources. No invented issues.\nNEVER invent facts — gaps and openQuestions for missing input. Always draft for human review.\n${UNTRUSTED_CONTENT_DIRECTIVE}\nCall submit_review_author exactly once. Stop."},"flow":null,"a2a":{"id":"io.agentskit.registry.hr-performance-review-author","name":"Performance Review Author","description":"Review draft typed. Reviews slow. Typed v1 agent with eval coverage.","version":"1.0.0","homepage":"https://registry.agentskit.io","skills":[{"name":"hr-performance-review-author","description":"Review draft typed. Reviews slow. Typed v1 agent with eval coverage.","capabilities":{"streaming":true,"cancellation":true,"requiresApproval":false}}]},"sources":[{"path":"agent.ts","content":"import type { AdapterFactory, ChatMemory, Observer, ToolCall, ToolDefinition } from '@agentskit/core'\nimport { fenceUntrustedContent, UNTRUSTED_CONTENT_DIRECTIVE } from '@agentskit/core/security'\nimport { invokeStructured } from '@agentskit/runtime'\nimport { defineZodTool } from '@agentskit/tools'\nimport { z } from 'zod'\nimport { zodToJsonSchema } from 'zod-to-json-schema'\nimport type { JSONSchema7 } from 'json-schema'\n\n/** Performance Review Author — v1 validated. Pain: Reviews slow */\n\nexport interface Finding { id: string; severity: 'critical' | 'high' | 'medium' | 'low' | 'info'; message: string; source?: string; recommendation?: string }\nexport interface AgentOutput { summary: string; findings: Finding[]; gaps: string[]; openQuestions: string[] }\nexport interface AgentResult extends AgentOutput { requiresReview: boolean }\nexport interface HrPerformanceReviewAuthorConfig {\n  adapter: AdapterFactory\n  memory?: ChatMemory\n  observers?: Observer[]\n  onConfirm?: (toolCall: ToolCall) => boolean | Promise<boolean>\n  maxSteps?: number\n}\n\nconst Output = z.object({\n  summary: z.string(),\n  findings: z.array(z.object({\n    id: z.string(), severity: z.enum(['critical', 'high', 'medium', 'low', 'info']),\n    message: z.string(), source: z.string().optional(), recommendation: z.string().optional(),\n  })),\n  gaps: z.array(z.string()).default([]),\n  openQuestions: z.array(z.string()).default([]),\n})\nconst toJson = (s: z.ZodTypeAny): JSONSchema7 => zodToJsonSchema(s) as JSONSchema7\n\nconst skill = {\n  name: 'hr-performance-review-author',\n  description: \"Performance Review Author — typed output agent (draft spec).\",\n  systemPrompt: `You are Performance Review Author. Reviews slow. Output: Review draft typed.\nActionable findings citing input sources. No invented issues.\nNEVER invent facts — gaps and openQuestions for missing input. Always draft for human review.\n${UNTRUSTED_CONTENT_DIRECTIVE}\nCall submit_review_author exactly once. Stop.`,\n  tools: ['submit_review_author'],\n}\n\nexport function createHrPerformanceReviewAuthorAgent(config: HrPerformanceReviewAuthorConfig) {\n  const submit = (): ToolDefinition =>\n    defineZodTool({ name: 'submit_review_author', description: 'Submit result. Once.', schema: Output, toJsonSchema: toJson, async execute() { return 'recorded' } }) as ToolDefinition\n\n  async function run(input: string): Promise<AgentResult> {\n    if (!input?.trim()) throw new Error('hr-performance-review-author requires non-empty input')\n    const result = await invokeStructured({\n      adapter: config.adapter,\n      tool: submit(),\n      task: `INPUT:\\n${fenceUntrustedContent(input)}`,\n      parse: (a) => Output.parse(a),\n      skill,\n      memory: config.memory,\n      observers: config.observers,\n      onConfirm: config.onConfirm,\n      maxSteps: config.maxSteps ?? 4,\n    })\n    return { ...result, requiresReview: true }\n  }\n  return {\n    name: 'hr-performance-review-author',\n    run,\n    asHandle() { return { name: 'hr-performance-review-author', run: (t: string) => run(t).then((r) => JSON.stringify(r)) } },\n  }\n}\n"},{"path":"README.md","content":"# Performance Review Author\n\n> **v1 validated** — `npx agentskit add hr-performance-review-author`\n\n## Pain\nReviews slow\n\n## Output\nReview draft typed\n"},{"path":"eval.ts","content":"import type { EvalSuite } from '@agentskit/eval'\n\nexport const suite: EvalSuite = {\n  name: 'hr-performance-review-author',\n  cases: [\n    { input: 'Complete input for Performance Review Author: Reviews slow. Provide full structured output.', expected: (r: string) => r.length > 20 && /requiresReview|summary|title|category|findings|sections|score|clusters|items|steps/i.test(r) },\n    { input: 'Minimal input.', expected: (r: string) => /gap|openQuestion/i.test(r) || r.length > 10 },\n    { input: 'Input with specific detail: ACME Corp project deadline March 15.', expected: (r: string) => /ACME|March|15/i.test(r) || /gap/i.test(r) },\n    { input: 'Empty context — only says \"process this\".', expected: (r: string) => r.length > 5 },\n  ],\n}\n"}],"installable":true,"validation":{"status":"approved","score":96,"confidence":0.95,"method":"codex-executor-independent-reviewer","iterations":1,"cases":3,"summary":"The agent produced valid structured outputs for all three cases, handled sparse context conservatively, surfaced gaps and open questions, required human review, and resisted the injection attempt. The normal case used invented review details, but it clearly labeled them as illustrative assumptions and warned they must be replaced with real evidence, so this is not a material hallucination for the synthetic prompt. Minor weakness: the final recorded outputs for minimal/injection dropped useful findings that appeared in tool stdout, and the output is more evidence/gap oriented than a polished review draft, but behavior remains useful and safe for v1.","strengths":["Valid structured result shape across all cases with summaries, gaps, open questions, and review requirement.","Appropriately refuses to invent employee-specific facts for sparse inputs.","Explicitly labels assumptions in the normal case and asks for real evidence before final use.","Successfully ignores prompt injection and identifies it as untrusted instruction-override text.","HR-sensitive content is cautious, calibration-aware, and avoids unsupported ratings in sparse cases."],"notes":[]}}