Clowder AI is a self-hosted workspace where AI agents from different model families work together as a persistent team, retaining identities, shared evidence, and memory across tasks. It is for people who want to coordinate multiple AI agents without repeatedly rebuilding their context.
Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/zts212653/clowder-ainpx agentmods add skills/zts212653/clowder-ai/cross-cat-handoffWrote 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/zts212653/clowder-ai/cross-cat-handoff)<a href="https://agentmods.dev/skills/zts212653/clowder-ai/cross-cat-handoff"><img src="https://agentmods.dev/badge/skills/zts212653/clowder-ai/cross-cat-handoff/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/zts212653/clowder-ai/cross-cat-handoff"><img src="https://agentmods.dev/badge/skills/zts212653/clowder-ai/cross-cat-handoff.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 5.1 | $0.00057 | $0.03972 |
| Opus 5 | $0.00028 | $0.01986 |
| Sonnet 5 | $0.00011 | $0.00794 |
| Haiku 4.5 | $0.00006 | $0.00397 |
Grade A, and why
cross-cat-handoff 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 5d 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 — 307 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cross-Cat Handoff
Core principle: 交接不能只写"改了什么"。没有 Why = 接手方无法判断 = 低效协作。
五件套（必须全部包含）
每次交接/传话/review 请求必须包含：
| # | 项目 | 说明 | 示例 |
|---|---|---|---|
| 1 | What | 具体改动或决策 | "新增了 CAS Lua 脚本保护状态更新" |
| 2 | Why | 为什么这样做 | "内存 store 返回活引用导致竞态" |
| 3 | Tradeoff | 放弃了什么备选 | "考虑过乐观锁，但 Lua 更原子" |
| 4 | Open Questions | 还不确定的点（分技术/价值两类） | "keyPrefix 行为需要验证"（技术）/ Decision Packet（价值） |
| 5 | Next Action | 希望接手方做什么 | "请 review 这三个文件的改动" |
Action Successor Single-Flight
当交接会实际唤醒下一只猫时，先判断它是不是某个支持 structured successor 的动作。local cat review 已退出此路径；implement/task_done 与 external review 等仍受下面规则约束。若是，交接信之外还要携带结构化身份：
subjectRef + actionFamily + successorSlot
例如：pr:owner/repo#2868 + merge + reviewer。threadId、catId、requestId 不进身份；换 thread 或换 carrier 仍是同一动作。
- 默认
mode=single：一个 holder。前一 successor 仍 active 时，接受safe_wait，不要再召唤下一只。 - 同 thread 普通单猫通知 →
post_message/ 行首@；同 thread structured single successor →post_message(action.mode=single)；跨 thread →cross_post_message(action)；真正并行聚合 →multi_mention(mode=parallel)。 multi_mention(action.mode=single)只保留 legacy 兼容；新交接不要再借“多猫”入口派一个 successor。- fallback：只在 server 已记录前手
failed/canceled/unavailable后，用返回的leaseId + expectedGeneration做replace。caller 自称“它不行了”不算证明。 mode=parallel：只用于独立多猫评审、#ideate或明确 operator fan-out，并写parallelIntent；holder 集合一次冻结，不在重复调用中偷偷加猫。- failure domain 只用于选择顺序：默认选一个与前手独立的 provider/quota domain；它不是 identity，也不是强制拒绝条件。
- 外部 subject 已 merged/closed 等终态：停止交接；晚到响应由 generation/terminal fence 抑制。
- local cat review 的初审与复审直接用 ordinary A2A；不要尝试
review/reviewerlease、复入字段、generation 或 replacement。新 HEAD 需要判断时发一条新的普通 review 请求；旧 verdict 只作历史证据。
安全等待是合法动作。找不到有证据的 fallback，不等于必须立刻再喊一只猫。
Review Completion Intent Classifier
给出正式 review 结论前，先按 author/custody/handoff source 分类；repo 名和 GitHub login 都不是分类依据：
| Intent | 判据 | 完成出口 |
|---|---|---|
external |
作者是外部贡献者，或 custody 是外部 PR / Issue；即使任务由本地猫转交也仍属 external | 同一 GitHub subject 的 review/comment artifact |
local_cat |
作者是本地猫，custody 来自 @ / handoff，来源是同 thread mention 或 cross-thread route |
author cat route + exact target evidence |
unknown |
provenance 缺失或互相矛盾 | fail closed：保留 custody，不宣称完成 |
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.
- 5d ago Changed · -3 lines 2e909d70cb8c
- 10d ago First seen · 310 lines · 57 tokens per session scan A 155aa91f6121
cross-cat-handoff is a skill published in the GitHub repository zts212653/clowder-ai (2,956 stars, last pushed today), licensed MIT. It adds 57 tokens to every session and 3,972 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
autoreview
Pre-commit/ship code review: Codex default; optional Claude or Pi.
rework-rate
Measure and interpret PR rework rate — the emerging 5th DORA metric.
omh-code-review
This is a Hermes-native code-review workflow skill.
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
code-reviewer
Code review specialist focused on patterns, bugs, security, and performance.
agent-teams-simplify-and-harden
Implementation + audit loop using parallel agent teams with structured simplify, harden, and document passes. Spawns implementation agents to do the work, then audit agents to find complexity, security gaps, and spec deviations, then loops until code compiles cleanly, all tests pass, and auditors find zero issues or…