agent-workflow-playbook

Solid

AI Agent Workflow & Skill Architecture Guide — turn expert work into measurable, reusable agent systems. Covers workflow discovery, skill decomposition, harness design, evaluation, human escalation, observability, cost control, and multi-agent orchestration. Includes a measured marketing-delivery case: 15 people × 3–4 weeks reduced to one strategist + AI in 5 days. By Gingiris.

AI & Automation 71 stars 5 forks Updated 2 days ago MIT

Install

View on GitHub

Quality Score: 88/100

Stars 20%
62
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
100
Issue Health 10%
50
License 10%
100
Description 5%
100

Skill Content

# AI Agent Workflow Playbook — 从专家经验到可规模化交付 > 适用于:把研究、营销、运营、分析、内容生产等高认知任务,改造成可测量、可纠错、可复用的 Agent 工作流。 ## 先判断:这个任务该不该 Agent 化 只有同时满足以下多数条件才进入自动化: - 输入和合格输出可以被描述; - 专家能说清“什么是对、什么是错”; - 任务重复发生,或交付成本随客户数近似线性增长; - 关键数据能合法、稳定取得; - 错误可以在发布、付款、删除或对外发送前被拦截; - 结果能通过 rubric、样例集或业务指标复核。 如果任务低频、目标持续变化、没有验收口径,先做人工 SOP,不要先搭多 Agent。 ## 1. 建立基线,不要直接写 Prompt 选择 10–30 个近期真实任务,记录人工基线: | 指标 | 定义 | |---|---| | 任务成功率 | 首次交付通过验收的任务数 / 总任务数 | | 一次通过率 | 无返工即通过的任务数 / 总任务数 | | 周期 | 从收到完整输入到可交付输出的 elapsed time | | 人工工时 | 研究、制作、复核、返工所花人时 | | 单次成本 | 模型、工具、数据和人工复核成本之和 | | 重试率 | 发生工具重试或整段重做的任务占比 | | 严重错误率 | 错误发布、错误付款、数据泄露等高风险事件占比 | 没有这张基线表,就只能证明 Agent “能跑”,不能证明工作流变好了。 ## 2. 从业务链路拆 Skill 先画业务链路,再按可验收结果拆 skill: ```text 需求澄清 → 数据获取 → 证据整理 → 分析 → 产出 → 质检 → 人工批准 → 交付 → 反馈沉淀 ``` 每个 skill 至少包含: ```yaml name: competitor-evidence-pack input_contract: required: [product, market, competitors, time_window] output_contract: required: [claims, source_urls, captured_at, confidence, unknowns] tools: allow: [search, fetch] deny: [publish, delete, payment] acceptance: - every material claim has a source - source capture time is recorded - unknown facts are labeled, not guessed escalate_when: - authenticated source is inaccessible - sources conflict on a decision-critical fact ``` 优先做单一职责 skill。只有当步骤间存在清晰依赖时,才增加 orchestrator。 ## 3. Harness:让系统知道边界、记住纠错、持续评测 Prompt 只描述一次交互;harness 管理长期运行环境。至少包含五层: 1. **Context**:品牌、客户、目标、禁区和数据权限; 2. **Skills**:通用技能与客户专属技能分离,按任务选择调用; 3. **Memory**:只沉淀经过确认的偏好、错误...

Details

Author
Gingiris-1031
Repository
Gingiris-1031/gingiris-skills
Created
2 months ago
Last Updated
2 days ago
Language
Python
License
MIT

Bundled in these plugins

Similar Skills

Semantically similar based on skill content — not just same category

AI & Automation Solid

multi-agent-coordinator

Use when executing implementation plans with independent tasks - dispatches fresh subagent for each task with code review between tasks, enabling fast iteration with quality gates. Supports orchestrator, peer-to-peer, and pipeline coordination modes.

47 Updated 2 days ago
huangwb8
AI & Automation Listed

ai-work-assetization-diagnoser

AI 工作资产化诊断器 / Assetization Router:当用户提供 AI 协作会话、重复任务、prompt、工作流描述、 团队 AI 使用场景或一次成功交付过程,并要求判断应该沉淀成 Prompt、Context Pack、Workflow、Skill、 Loop、System,或根本不值得沉淀时使用。它只做诊断、分层和下一步资产建议,不替代具体 Skill 创建、 PRD 起草、调研、实现或自动化执行。不用于普通事实查询、一次性文案、trace 根因分析或高责任专业判断。

5 Updated 5 days ago
PANGKAIFENG
AI & Automation Listed

agent-skill-design

Design, review, or refine the behavioral substance of Agent Skills: decide whether a repeated task belongs in a Skill, define its responsibility, invocation and non-invocation conditions, degree of freedom, tool and resource boundaries, failure handling, and evaluation cases. Use when turning workflows or guidance into a Skill design or correcting a Skill that is vague, broad, unsafe, hard to invoke, or difficult to evaluate. Do not use for platform-specific packaging, manifests, file templates, installation, or authoring-spec compliance.

1 Updated 2 weeks ago
hudrazine