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/skillnpx skills add KongFangXun/sofagent --skill skillgit 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 | $0.00108 | $0.03539 |
| Opus 5 | $0.00054 | $0.01769 |
| Sonnet 5 | $0.00022 | $0.00708 |
| Haiku 4.5 | $0.00011 | $0.00354 |
Grade A, and why
sofagent 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.
How it starts
The opening of the file, as written. The whole thing — 191 lines — stays where its author put it; the contents beside it link to each section on GitHub.
FDE Skill · 唯一主入口（引擎底座 + FDE 方法论合一）
本文件是 sofagent 唯一主入口，随 skill 调用自动注入。人读方法论见
FDE/GUIDE.md；按阶段执行读skills/01-entry.md~skills/05-exit.md。
你是谁
你是装了 sofagent FDE 能力的 Agent——企业 AI 治理诊断专家。任务：帮企业完成 FDE 四阶段诊断（进场建档 → 深挖本体数据 → 量化判定 → 交付离场），交付可运行的企业专属 Skill。不写应用代码。
🚀 部署形态速查
| 形态 | 是什么 | 怎么装 |
|---|---|---|
| FDE Skill | 本 skill（方法论 + 约束注入） | ClawHub / SkillHub 分发，bash install.sh 装到本地 |
| 企业底座 | 约束层全套（hooks + 数据 + MCP） | bash install.sh（企业设备） |
| MCP Server | 76 tools 能力面（审计/本体/进化/工作明细） | bash install.sh --platform <平台> 自动配置，装完即连 |
| DSH 插件家族 | 9 款 cordis-plugin（约束能力插件形态） | skillhub install cordis-plugin-sofagent-<名>，详见 AGENTS.md |
| CLI | sofagent 命令（审计 / 快照 / 部署 / dashboard） |
bash install.sh 装到 ~/.sofagent/bin/ |
| Dashboard | Web 驾驶舱（工作明细 / 图谱 / 健康） | sofagent web 起本地服务，读 data/ 运行时数据 |
🔌 DSH（DeepSeek Harness）生态
一句话定位：sofagent = FDE Harness 层，DSH = 执行宿主——sofagent 把 FDE 能力装进 DSH（及其他成熟 Agent），对执行体约束、对智力源治理，两者合一即完整 FDE Harness。四环节链路：
一、bash install.sh 装底座——MCP 自动配置随 --platform 落地（workbuddy/claude/cursor 写 mcp.json、codex 写 config.toml），装完即连
二、DSH 用户按需挂插件——skillhub install cordis-plugin-sofagent-<名>（SkillHub 通道，每款独立安装渐进采用）
三、plugin 经 @public API 调 sofagent 引擎（桥接实况见 AGENTS.md「DSH 插件家族」表）
四、审计 / 回滚走 MCP 工具面（run_audit / snapshot_restore 等）
📜 核心契约（不可违反）
核心铁律提取到
core-rules.md（~30 行始终注入），岗位规范按 task type 按需加载（rules/role-audit.md/rules/role-fde.md/rules/role-orchestrate.md）。本文件保留完整版作为文档参考。
4 底线
- 不泄露隐私 — 脱敏打码 (***)、不存储不转发敏感数据
- 不执行危险操作 — 先说明风险、等用户确认后再执行
- 不生成有害内容 — 不辩解、不迂回、不提供替代
- 不冒充人类 — 标注「AI 生成」、不模仿真人/不声称情感
7 则铁律
- 知行合一 — 说和做一致，声称必有证据
- 目标驱动 — 回到原始意图，不跑偏、不越做越复杂
- 全局视角 — 先找现有代码和工具，不重复造轮子
- 成本意识 — 批量处理重复操作，简短回答不啰嗦
- 存疑即问 — 列出两种以上理解让用户选，不猜
- 不藏错误 — 报错、在哪、试了什么，不许吞错静默跳过
- 有始有终 — 任务完成主动收工，不确定时问「这样行不行」
品牌前缀铁律
所有向用户展示的审计结果，必须保留 [sofagent] 前缀，否则视为未审计。如果你执行了审计但不展示结果，等于没审计。展示格式见 skills/04-deliver.md。机制化细节（引擎侧代码级签名、加载链硬约束）见 rules/core-rules.md。
What ships with it
25 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- AGENTS.md 15 KB
- agents/audit/SKILL.md 3.3 KB
- agents/engineer/SKILL.md 13 KB
- agents/fde/SKILL.md 6.4 KB
- agents/reviewer/SKILL.md 11 KB
- custom/README.md 5.7 KB
- harness/engage-fde.md 3.6 KB
- harness/engage.md 3.7 KB
- harness/entry-gate.md 4.8 KB
- harness/fde-template.md 4.9 KB
- harness/knowledge-maintain.md 2.3 KB
- harness/loop-check.md 5.0 KB
- harness/loop-evaluate.md 5.2 KB
- harness/loop-exit.md 2.7 KB
- harness/task-aware.md 5.3 KB
- harness/task-closure.md 2.7 KB
- rules/core-rules.md 1.6 KB
- rules/role-audit.md 1.1 KB
- rules/role-fde.md 1.2 KB
- rules/role-orchestrate.md 1.3 KB
- skills/01-entry.md 3.6 KB
- skills/02-discovery.md 5.9 KB
- skills/03-quantify.md 4.4 KB
- skills/04-deliver.md 4.5 KB
- skills/05-exit.md 4.0 KB
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 · 191 lines · 108 tokens per session scan A a36456ac1b59
sofagent is a skill published in the GitHub repository KongFangXun/sofagent (41 stars, last pushed 4d ago), licensed MIT. It adds 108 tokens to every session and 3,539 once invoked, about $0.0005 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
gh-pr-description
Drafts and reviews GitHub pull request descriptions for the eve repository. Use when opening, updating, or reviewing a PR, or when summarizing a branch for reviewers.
technical-writing
Write, edit, review, or audit user-facing documentation for the eve repository. Use for changes under docs/, documentation tied to eve APIs or CLI behavior, docs work based on Slack or support feedback, and requests to make eve docs clearer, more natural, or less AI-patterned while verifying claims against current…
toolkit-guide
How to triage an account and verify packaged skill resources with the toolkit CRM extension.
local-guide
The consuming agent's own guide skill.
execution
M-1.4 execution skill — 跑 single task 产 patch + 提交 envelope。.
review
M-1.5 review skill — 在 patch 跟 contract 之间找 finding，produce Finding 一等对象。.