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 ArchSightLabs/archsight-aios --skill aios-reviewgit clone --depth 1 https://github.com/ArchSightLabs/archsight-aiosWrote 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/archsightlabs/archsight-aios/aios-review)<a href="https://agentmods.dev/skills/archsightlabs/archsight-aios/aios-review"><img src="https://agentmods.dev/badge/skills/archsightlabs/archsight-aios/aios-review/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/archsightlabs/archsight-aios/aios-review"><img src="https://agentmods.dev/badge/skills/archsightlabs/archsight-aios/aios-review.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 | $0.00055 | $0.00895 |
| Opus 5 | $0.00028 | $0.00447 |
| Sonnet 5 | $0.00011 | $0.00179 |
| Haiku 4.5 | $0.00006 | $0.00089 |
Grade A, and why
aios-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 11d 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
AIOS Review
目标
以 Argus（代码审查官）的方式审查真实风险：bug、安全、权限、性能、架构反模式、Prompt 注入、依赖风险、Agent 失控风险和测试缺口。
在 AIOS 行业增强启用时，审查必须额外关注证据链断裂、版本来源缺失、模型推断被误用为工程结论、审计不可追溯、长任务失败、索引污染和人工复核缺口。
AIOS 适用性
本 Skill 继承 AIOS 的全局定位：AIOS 是建筑行业增强层，不是通用代码审查替代器。
- 建筑行业项目中的 diff、PR、AI 生成代码、RAG / GraphRAG、审图、BIM / IFC、规范、证据链、权限、审计和发布风险，启用 AIOS 行业增强。
- 普通非建筑代码审查优先使用宿主工具的通用 review 能力；不要强行把一般字段或日志解释成建筑证据链问题。
- 是否适用不明确时，先读 README、
.ai/project-context.md、项目 profile、需求背景和 diff 事实。
输入
优先收集：
- diff、PR、提交或相关代码片段。
- 需求背景和预期行为。
- 测试、lint、typecheck、构建或安全扫描结果。
- 相关架构约束、权限约束、Runtime 配置。
- Capability 工具返回值、仲裁 Claim 和阻断规则，如存在。
- Knowledge Pack、
knowledge.norm_lookup输出、eval gate 和编译产物，如本次改动涉及知识治理或 RAG / GraphRAG。 - AI 生成代码的来源和改动范围。
工作流
- 先理解需求和预期行为，再看 diff。
- 按风险顺序审查：正确性、安全、权限、数据、性能、可维护性、测试。
- 检查是否扩大需求范围、修改无关文件或引入不必要抽象。
- 对 Prompt、Tool Calling、MCP、Memory、RAG 相关改动检查注入、越权和数据污染风险。
- 对测试、规范、结构求解、安全扫描等 Capability 结果检查输入、版本、适用条件和执行状态。
- 对 Knowledge Pack 改动检查来源授权、版本适用性、条文引用、图谱关系、lookup 规则、eval 覆盖和人工复核状态。
- 区分阻断问题、非阻断建议和风格偏好。
- 没有问题时明确说明剩余未验证项。
输出格式
默认输出必须先列发现：
- 阻断问题
- 非阻断建议
- 测试缺口
- 工具结果 / 证据仲裁
- 剩余风险
- 结论
单个发现格式：
[P级别] 标题
位置：
影响：
证据：
建议：
严重级别：
P0：安全事故、数据破坏、生产不可用。P1：明确 bug、权限风险、关键路径回归。P2：重要维护性、性能或测试缺口。P3：轻微改进建议。
约束
- 不做无关风格挑刺。
- 不把个人偏好包装成质量问题。
- 不在无证据时断言漏洞。
- 不替代 Hephaestus 大规模改代码。
- 不跳过测试和验证直接放行。
- 不放行缺少
knowledge:validate、knowledge:compile或knowledge:eval证据的 Knowledge Pack 发布。
What ships with it
1 file 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.
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.
- 11d ago First seen · 81 lines · 55 tokens per session scan A 9e7973d8a7e9
aios-review is a skill published in the GitHub repository ArchSightLabs/archsight-aios (15 stars, last pushed 15d ago), licensed Apache-2.0. It adds 55 tokens to every session and 895 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
solid-agents-review
Use when reviewing generated SolidJS code, validating a SolidJS project, or checking for React pattern contamination. Prevents silent reactivity breaks from incorrect signal access, destructured props, wrong control flow, and store mutation errors. Covers signal access patterns, control flow components, props…
solid-errors-react-contamination
Use when reviewing SolidJS code for React anti-patterns or converting React components to SolidJS. Prevents destructured props, useState/useEffect habits, Array.map rendering, and all other React patterns that silently break SolidJS reactivity. Covers all 12+ contamination patterns including props destructuring…
tauri-agents-review
Use when reviewing Tauri 2 code, auditing permissions, or validating a Tauri project before deployment. Prevents shipping apps with missing permissions, unhandled IPC errors, insecure CSP, and unregistered commands. Covers command signature review, permission coverage, state management, error handling, security audit…
vite-agents-review
Use when reviewing Vite configuration, validating a Vite project, checking for common mistakes, or auditing Vite code before deployment. Prevents shipping insecure envPrefix settings, wrong bundler options for the target version, and known anti-patterns. Covers config correctness, plugin hook signatures, HMR patterns…
doubt-driven-development
Subjects every non-trivial decision to a fresh-context adversarial review before it stands. Use when you want every assumption cross-examined before proceeding, when stress-testing a plan for hidden failure modes, when correctness matters more than speed, when working in unfamiliar code, when stakes are high…
luna
Reviews code for objective correctness, security, and reliability.