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
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 skills add zts212653/clowder-ai --skill request-reviewgit clone --depth 1 https://github.com/zts212653/clowder-aiWrote 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/request-review)<a href="https://agentmods.dev/skills/zts212653/clowder-ai/request-review"><img src="https://agentmods.dev/badge/skills/zts212653/clowder-ai/request-review.svg" alt="Measured on agentmods" 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 | $0.00089 | $0.04002 |
| Opus 5 | $0.00044 | $0.02001 |
| Sonnet 5 | $0.00018 | $0.00800 |
| Haiku 4.5 | $0.00009 | $0.00400 |
Grade A, and why
request-review 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 — 231 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SOP definition:
sop-definitions/development.yamlstagereview。
Request Review
把当前 diff、最高风险面和真实验证证据送到一只非作者猫眼前。默认只选一个合适的独立验证源；local peer、cloud、愿景守护各按自己的风险触发，不能因为“进入 review”就自动叠加。
先选验证源
| 风险需要 | 独立验证源 |
|---|---|
| 家里语境、skill/SOP/治理文字、实现语义 | local peer（本 skill） |
| 安全 / 鉴权 / 生产数据 / 外部契约，或需要 context-blind 代码扫描 | cloud；不再同时把同一问题默认发 local |
| 用户可见 feature 的终态是否符合愿景 | 愿景守护；在 feature close 触发，不是每个 PR 的 reviewer |
安全、数据或契约高风险需要不同视角时可以叠加；叠加理由必须指向不同风险面。相同目的的重复 reviewer 不增加门禁强度，只增加等待。
选择边界：只有新增了需要第二只猫判断的实质内容，或风险面明确要求独立验证时才进入本 skill；“有 diff”“SHA 变了”“开了 PR”都不是独立触发器。机械登记、已审内容转录、低风险 direct-main docs 与可证明 continuity 可以 skip/reuse。一旦选择 review，同一个体不能 review 自己，证据须覆盖最终实质内容（exact SHA 或 continuityProof）。
稀缺判断席位（dossier-driven）
当队友 dossier / L0 把 reviewer 标为周额度稀缺的高杠杆判断猫（当前为 Fable）时，不能沿用普通迭代 reviewer 的默认回路。请求必须写明
engagementMode=one_shot_calibration|final_seal、本轮唯一判断问题、停止条件和修后去向：
one_shot_calibration：用于重要 plan、架构/failure-mode 校准。reviewer 一次性交付判断和 findings 后退出；作者负责修复与测试，仍需独立验证时转日常 reviewer。final_seal：用于其他猫已经完成多轮审查、分歧和证据已整理后的最终封版讨论。只有 scope 已稳定、预期不再进入实现陪练时才选。- 普通修复不复入原稀缺 reviewer。P1/P2 的严重度不自动等于需要同一只高杠杆猫再次判断；只有修复引入新的架构/决策问题、原 finding 无法机械验收，或 operator 明确要求时才可复入。
稀缺指的是可用判断席位，不是单 token 价格。省下来的额度应留给“判断错的代价高、验证器弱”的节点；routine fix、exact-HEAD 续签、礼貌确认都不占这个席位。
发请求前
| 证据 | 何时必需 | 缺失动作 |
|---|---|---|
| 当前 diff / branch / HEAD | 始终 | BLOCKED — reviewer 不审漂移目标 |
Review-Subject-Ref + accepted source/revision |
始终 | BLOCKED — 无法把 verdict 归到同一 review episode 或判断 source 是否移动 |
| 五轴风险判断（行为、数据、安全、契约、不可逆） | 始终 | BLOCKED — 无法判断 review 深度 |
| 与风险匹配的验证输出 | 始终 | BLOCKED — 先跑 targeted 或 full gate |
| 原始需求摘录 | 涉及用户意图 / 愿景 | BLOCKED — reviewer 无法判断做没做对 |
| Architecture cell / Map delta / Why | 结构或 ownership 变化 | BLOCKED — 回设计面补齐 |
| author 浏览器 preview 记录 | 前端行为 / 视觉变化 | BLOCKED — author 先实际打开页面验证 |
| 根目录工件闸门 | 有媒体 / 设计证据 | BLOCKED — 先归档或移出仓库根 |
前端证据边界
前端验证劳动属于 author：自己启动/复用正确的 preview，走一遍关键交互，并记录 URL、操作和结果。截图、录屏、DOM assertion、Playwright 输出都可以作为证据载体；缺截图本身不是 operator 补劳动的许可证。
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 Changed · +44 lines 6c5c2d2e77c2
- 8d ago First seen · 187 lines · 89 tokens per session scan A 687149bb5888
request-review is a skill published in the GitHub repository zts212653/clowder-ai (2,924 stars, last pushed today), licensed MIT. It adds 89 tokens to every session and 4,002 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 2026-08-30.
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…