Wanwu is an enterprise platform for building AI agents, workflows, retrieval-augmented applications, and managing models in multi-tenant environments. It is designed for developers and enterprise teams delivering AI applications and integrations. The catalogue entries provide skills and agents for using the platform.
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/unicomai/wanwu/archive-protocolnpx skills add UnicomAI/wanwu --skill archive-protocolgit clone --depth 1 https://github.com/UnicomAI/wanwuWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/unicomai/wanwu/archive-protocol)<a href="https://agentmods.dev/skills/unicomai/wanwu/archive-protocol"><img src="https://agentmods.dev/badge/skills/unicomai/wanwu/archive-protocol.svg" alt="Measured on agentmods" height="20"></a>What 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 | $0.00087 | $0.01361 |
| Opus 5 | $0.00044 | $0.00681 |
| Sonnet 5 | $0.00017 | $0.00272 |
| Haiku 4.5 | $0.00009 | $0.00136 |
Grade A, and why
archive-protocol 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 — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
全局归档协议
只要当前任务需要写入任何文件,就必须执行本协议。
归档物包括但不限于:PLAN.md、report.md、summary.md、result.json、notes.md。
【身份文件】
若任务需要人格/身份设定,只认:
SOUL.mdIDENTITY.md
未确认可读或已注入上下文前,不得声称「已成功读取」。
归档协议 (Archive Protocol)
【核心规则:必须全量归档】
所有生成的成果性内容(无论是单个文件还是整个目录/文件夹),在任务完成后必须彻底搬移(Move)至归档区。
工作区(Workspace)仅作为临时加工场,任务结束后的合规状态是:工作区不留任何生成物,全部进入 archives/。
【绝对路径硬性边界】
所有文件(包括工作区临时文件、PLAN.md、归档物、报告、JSON 等)的保存路径必须落在以下根目录之下,禁止越界:
/home/root/workspace/{WORKSPACE_ID}/workspace/
{WORKSPACE_ID}由当前会话上下文给定,由archive工具拼接,不得手填、不得使用其它前缀。- 工作区与归档区都必须以该绝对前缀开头:
- 工作区临时文件:
/home/root/workspace/{WORKSPACE_ID}/workspace/<相对路径> - 归档区文件:
/home/root/workspace/{WORKSPACE_ID}/workspace/archives/{ARCHIVE_ID}/...
- 工作区临时文件:
- 任何写入到该前缀之外的路径(如
/tmp/...、/root/...、./xxx、其它绝对路径)都视为越界,必须立即终止并报错;禁止用软链接、相对路径..跳出该根。 archive工具返回卡片中的archive_root必须以/home/root/workspace/开头;不满足则视为执行异常。
【归档路径规则】
路径执行双轨制,严禁混淆,并且只能由 archive 工具写入。注意:归档操作会将文件或目录从工作区物理搬移(Move)至归档区,原位置将不再存在。
- 计划文件(PLAN.md 专属):工具参数
{"kind":"plan","sourcePath":"PLAN.md"}。归档后,如需修改计划,应直接编辑archives/{ARCHIVE_ID}/PLAN.md(绝对路径:/home/root/workspace/{WORKSPACE_ID}/workspace/archives/{ARCHIVE_ID}/PLAN.md)。 - 普通归档物(单个文件或整个目录):工具参数
{"kind":"file","sourcePath":"result_dir"}。支持归档单个文件或包含多个文件的目录。
【目录归档模式】
当你的输出包含多个文件(如前端项目、代码包、多份分析报告)时,必须执行以下流程:
- 在工作区创建一个专用文件夹(如
output/),位置必须在/home/root/workspace/{WORKSPACE_ID}/workspace/之下。 - 将所有相关文件放入该文件夹。
- 调用
archive工具对整个文件夹进行归档(sourcePath指向文件夹路径)。 - 归档成功后,整个文件夹及其内容将移入
archives/{ARCHIVE_ID}/{TIMESTAMP}/(绝对路径:/home/root/workspace/{WORKSPACE_ID}/workspace/archives/{ARCHIVE_ID}/{TIMESTAMP}/)。
【写入后强制校验】
任何归档文件写入后,必须立即回读校验。未完成校验前,不得声称「已成功归档」。调度 archive 工具会在写入后自动回读并输出状态卡片。
输出
请在最终输出的结尾,把archive工具的输出卡片也输出出来。必须以markdown的json代码块输出。
【工具执行指引】
所有归档物生成都必须通过 archive 工具 完成;禁止直接写 archives/ 目录。推荐流程:
- 准备源文件:在工作区临时目录编写内容。
- 调用工具(工具会自动解析 session,生成
TIMESTAMP):- 单个文件归档:
{ "name": "archive", "arguments": { "kind": "file", "sourcePath": "result.json" } } - 整个目录归档(推荐用于多文件结果):
{ "name": "archive", "arguments": { "kind": "file", "sourcePath": "my_results_folder" } }
- 单个文件归档:
- 后续操作:归档成功后,工作区原文件/目录已删除。若需再次查看或编辑,必须使用工具返回的 JSON 卡片中的
subpath或archive_root拼接出的完整路径 访问归档区文件。 - 异常处理:工具返回错误提示(如
Archive blocked)时须立刻停止归档流程,引用错误原因告知用户并等待下一步指示;禁止在失败后自行补写文件或回执。
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 · 95 lines · 87 tokens per session scan A 267de1f53596
archive-protocol is a skill published in the GitHub repository UnicomAI/wanwu (2,456 stars, last pushed yesterday), licensed Apache-2.0. It adds 87 tokens to every session and 1,361 once invoked, about $0.0004 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 2026-09-03.
Other skills, from other repositories
Book2Skill
Convert one or more TXT, Markdown, DOCX, or PDF documents into a reusable skill zip backed by normalized Markdown, extracted images, and a grounded JSONL knowledge index. Invoke this skill before inspecting task files, then execute its workflow directly without listing directories.
google-workspace
Gmail, Calendar, Drive, Docs, Sheets via gws CLI or Python.
box
Box manages cloud files, sharing, search, and metadata.
apple-notes
Manage Apple Notes via the memo CLI on macOS (create, view, search, edit).
co-mail-and-drive
Read and send mail from the user's own Gmail or Outlook account, send from the agent's own address, manage their Outlook contacts, and list/search/download/upload their Google Drive files — with co gmail, co outlook, co email, and co gdrive. Use when the user asks about their inbox, an email they received or want to…
output-build-workflow
Implement an Output SDK workflow from a plan document. Use when the user asks to build, implement, or code a workflow from an existing plan, or after output-plan-workflow has produced a plan and the user is ready to build.