Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/kongfangxun/sofagent/skill-templatenpx skills add KongFangXun/sofagent --skill skill-templategit clone --depth 1 https://github.com/KongFangXun/sofagentWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00055 | $0.01052 |
| Opus 5 | $0.00028 | $0.00526 |
| Sonnet 5 | $0.00011 | $0.00210 |
| Haiku 4.5 | $0.00006 | $0.00105 |
Grade A, and why
enterprise-fde-识别AI节点 scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
企业专属 Skill · 识别 AI 节点（sofagent 自案例 · 模板）
📋 这是模板文件——给客户企业每个 AI 节点定制的 Skill 格式参考。 不是 FDE Harness 自己的 Skill（那个主入口在
SKILL/SKILL.md，阶段指引在SKILL/skills/01-05）。FDE 在 §7-§8 定制。每个 AI 节点一个专属 Skill，注入企业的行业术语 / 业务规则 / 历史案例。 实际产出路径：
skills/fde-识别AI节点/SKILL.md（每个节点一个目录，SKILL.md 是入口）。 初始版本由 FDE 写，运行后基于 eval.md 评分自动迭代——不需要 FDE 手动调。
行业术语
这个节点所在领域的专有词汇和缩写：
| 术语 | 含义 | 用在哪 |
|---|---|---|
| 🔄 自动执行 | 规则明确、高频重复、输入输出可结构化的节点 | 分类判定结果 |
| ⚡ 强化岗位 | 需要判断但规则可描述，AI 做领航员的节点 | 分类判定结果 |
| 👤 暂时不动 | 依赖直觉经验或数据不可达的节点 | 分类判定结果 |
| 五要素 | 输入/输出/负责人/耗时/痛点 | 节点完整性判断 |
| 三问判定法 | 输入能自动取？规则能写清？输出能自动推？ | 分类核心规则 |
业务规则
这个节点必须遵守的规则。Agent 会严格执行：
- 输入必须是完整的五要素——缺任何一项先标记"不完整"，不强行判定
- 三问判定法的答案必须基于事实（"有 API"≠"API 能用"），不能猜
- 边界情况（两问 yes 一问 no）判为 ⚡，不是 🔄
- 判定理由必须一句话写清楚，不能只填"符合 🔄 特征"
- 任何人都可以 override 判定结果，但 override 必须记录理由
历史案例
这个节点跑过的典型案例，Agent 做决策时参考：
案例 1：数据拉取 → 🔄
- 输入：ERP API（有接口，能自动取）
- 规则：拉昨日销售数据（新人照着文档能做）
- 输出：写入飞书表格（API 可推）
- 判定：三 yes → 🔄
案例 2：设计稿审核 → ⚡
- 输入：Figma 设计稿（能自动取）
- 规则：审核是否符合设计规范（规范可描述，但需要审美判断）
- 输出：审核意见推给设计师（能自动推）
- 判定：两 yes 一 no（规则需判断）→ ⚡
案例 3：战略决策 → 👤
- 输入：市场数据 + 老板直觉（直觉不可自动取）
- 规则：没有固定规则
- 输出：决策结果
- 判定：零 yes → 👤
约束
这个节点的 Skill 不准做什么：
- 不准在五要素不全时强行判定
- 不准跳过三问直接凭经验分类
- 不准把"理论上能做到"当成"实际能做到"（"有 API"但"API 没开权限"= 不能自动取）
- 不准修改业务流节点图（只读输入，只写分类清单）
迭代记录
Skill 优化分析自动维护，FDE 不用手动写：
| 版本 | 日期 | 改了什么 | 触发原因 |
|---|---|---|---|
| 1.0 | 2026-07 | FDE 创建初始版本 | — |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 85 lines · 55 tokens per session scan A d88dfe980e1f
enterprise-fde-识别AI节点 is a skill published in the GitHub repository KongFangXun/sofagent (41 stars, last pushed 4d ago), licensed MIT. It adds 55 tokens to every session and 1,052 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
agent-workflow-playbook
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…
handoff
当你正要把一件事交到另一个智能手里时用它——派发一个 worker、向上回报、问人/问 owner、为后来的 agent 留下上下文，或装配要喂给下游的 context。它装的不是"怎么写一段提示词"，是"一个智能该怎么跟另一个智能沟通"：先建对方的读者模型，只交思考和判断、把路留给对方，在控制与信任之间导航。只要你在写一段会被另一个 AI 读的文字——dispatch、brief、handoff、交接信、context 装配、给下游 agent 的指令——就用它，哪怕你只把它叫做"写个派发提示词"。正要派任务的时刻同样算：dispatch 一个 task、起一个 fork / worker / fix 会话 /…
b3os-team-member-lifecycle
Use when adding, testing, disabling, offboarding, or archiving a b3rys team member across Claude Code channel, OpenClaw, Hermes Agent, or future runtimes.
b3os-bwf
BWF(b3os workflow)는 harness가 아니다. b3rys 팀이 과제를 받아 끝까지 수행하는 기본 흐름이다. harness·multi-ai review·judge panel·적대적 리뷰는 이 흐름의 한 단계에서 고르는 품질방법일 뿐이다. (이 구분을 먼저 박는다 — 과거 "BWF=harness" 오해 방지.).
b3os-scheduler
팀원이 세션과 무관하게(durable) 시각에 맞춰 발화하는 반복/예약 잡을 등록하는 절차 스킬이다. 런타임별 세션 cron(claude CronCreate, openclaw/hermes 자체 예약)은 그 세션이 죽으면 유실된다. 팀 정본 반복작업(야간 사이클·주기 리마인드·정기 점검)은 이 스케줄러에 등록해 team.db에 durable 저장하고, b3os 서버 안의 워커가 발화한다.
b3os-harness-playbook
Skill "b3os-harness-playbook" from b3rys/b3rys-team-os, covering b3os-harness-playbook — harness 플레이북 (트리거 우선), ① 트리거 — 이 모양이면 기본 harness (보면 바로 레시피), ② ready-run 레시피 (모양별 — 골라서 바로 실행), ③ 구조 — 제대로 짜는 모양 (팀원이 하네스 짤 때 필독) and ④ 하드 캡 + 런타임별 실행.