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-arch-healthgit 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-arch-health)<a href="https://agentmods.dev/skills/archsightlabs/archsight-aios/aios-arch-health"><img src="https://agentmods.dev/badge/skills/archsightlabs/archsight-aios/aios-arch-health/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-arch-health"><img src="https://agentmods.dev/badge/skills/archsightlabs/archsight-aios/aios-arch-health.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 | $0.00137 | $0.01260 |
| Opus 5.5 | $0.00055 | $0.00504 |
| Sonnet 5.5 | $0.00027 | $0.00252 |
| Haiku 4.5 | $0.00014 | $0.00126 |
Grade A, and why
aios-arch-health 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 2mo 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 — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AIOS Arch Health
建立可复验的架构健康事实和棘轮门禁。不要用本 Skill 替代 aios-arch 的业务边界判断,也不要让自然语言推断伪装成扫描事实。
aios-arch-health 是唯一 canonical Skill ID。aios-architecture-health 和 archsight-architecture-health 只作为自然语言兼容触发,不建立重复 Skill。
分工
- 使用本 Skill 执行扫描、证据来源核验、受保护约束差分、基线差分、临时预算和门禁。
- 使用
aios-arch阅读业务与系统上下文,解释热点是深 Module、合理复杂度还是职责混杂,并评估服务、数据、模型和 Runtime 边界。 - 使用
aios-review消费 PR 中的 SARIF 和门禁结果。 - 使用
aios-plan把确认后的治理项拆成交付任务。
工作流
- 读取项目规则、
.ai/architecture-health/profile.json或用户指定的 profile。 - 选择运行模式:
commit:关注相对基线的新增债务,目标为数十秒级。weekly:运行完整扫描、趋势对比和热点排序。milestone:增加性能、真实数据库、并发或失败注入等 profile 要求的证据。
- 运行项目自有扫描器,或读取符合输入契约的归一化事实文件。
- 按 profile 收集 unit、acceptance、property、mutation、coverage、code quality、dependency、UI QA、performance、real database、concurrency 和 failure injection 等分层证据;不要让低成本证据冒充高风险验证。
- 当 profile 要求时,核验必需证据的 tool、command、actor、role、repository commit、observedAt、environment 和 artifact SHA-256。
- 对照基线识别受保护的 specification、acceptance test、unit test、quality profile 和 QA procedure 是否新增、修改或删除;变化必须有覆盖全部变更、对应当前提交且非生产者自审的
constraint-approval。 - 生成
architecture-health.json、Markdown、SARIF、依赖图和循环依赖清单。 - 只让
measured的确定性违规触发 fail;inferred只进入建议;证据或约束不可验证时 HOLD。 - 对照基线区分
new / worsened / improved / retained / resolved / budgeted。 - 校验临时预算的作用域、负责人、原因、上限和到期时间;过期预算不得放行。
- 把需要业务解释的热点交给
aios-arch,同时保留原始事实和证据定位。
CLI
优先使用仓库提供的命令:
archsight-aios architecture:health \
--cwd <project> \
--mode <commit|weekly|milestone> \
--health-profile <profile.json> \
--health-input <facts.json> \
--baseline <previous-architecture-health.json> \
--out <artifact-directory>
项目可以在 profile 中声明本地 analyzer command。不要把 Compliance、Standards、Graphics 的阈值或扫描命令写入 AIOS 核心;由各项目在独立 Goal 中维护自己的 profile、适配器和基线。
证据规则
measured:确定性工具或可复验运行结果。inferred:由事实推导的架构风险,不能直接阻断。unverified:profile 要求但本次没有取得的证据。- 必需证据开启 provenance 后,必须绑定当前 repository commit;需要 artifact 时必须提供仓库相对路径和小写 SHA-256。
- 受保护约束有变化时,必须由
reviewer提供 measured 的constraint-approval,覆盖所有变更约束;约束生产者不得自审。 - 代码行数、扇入和扇出只作调查信号,不单独作为删功能或拆模块的硬门禁。
- 性能结果必须携带环境、数据集和运行条件。
- 静态事务或并发信号不能冒充真实数据库、竞态或失败注入结论。
- SARIF 承载可定位的确定性发现;AI 解释保持 advisory。
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.
- 2mo ago First seen · 73 lines · 137 tokens per session scan A e594dcb88d52
aios-arch-health is a skill published in the GitHub repository ArchSightLabs/archsight-aios (27 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 137 tokens to every session and 1,260 once invoked, about $0.0005 per session on Opus 5.5. 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
agent-qa-debug-fix
Debug, patch, and verify failed Agent QA runs from MCP evidence, artifacts, logs, and local code without hiding product or infrastructure defects.
agent-qa-result-triage
Triage failed Agent QA runs with MCP evidence, artifacts, logs, fixed failure categories, confidence, and actionable next steps.
autonomous-tdd-debugger
Empowers the agent to autonomously run tests, read terminal stack traces, and self-heal code until tests pass. Transforms the agent from a passive coder to an active CI pipeline debugger.
agent-qa-run-review
Review recorded agent-qa runs and memidex provenance to explain a QA verdict, identify missing evidence, and propose the smallest next step.
bug-fix-protocol
8-step disciplined bug-fix protocol that treats every production bug as two failures — the code defect itself and the testing system that allowed it through. Use when fixing a production bug, investigating a regression, writing a post-mortem, or auditing a missed defect. Triggers on "fix this bug", "production bug"…
polya-fix
Phase 5 of the Pólya Heuristic Coder: First-Principles Bug Remediation, Seam Tracing, Bisection Search (O(log n)), Failing Reproduction Tests, and Incubation Circuit-Breaker Gate (docs/bug-reports/{slug}-bugfix.md).