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 commands/alisunstar/opensunstar/extract-routinggit clone --depth 1 https://github.com/alisunstar/OpenSunstarWhat 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 | $0.00000 | $0.00608 |
| Opus 5 | $0.00000 | $0.00304 |
| Sonnet 5 | $0.00000 | $0.00122 |
| Haiku 4.5 | $0.00000 | $0.00061 |
Grade A, and why
extract-routing 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 yesterday.
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
/rd:extract-routing — 源码锚点提取适配器（外置）
知识资产层适配器：从源码提取业务锚点（Topic/状态枚举/业务身份常量/接口名/表名），产出锚点补表草稿入 knowledge/candidate/，经 owner review 后才并入正式页 frontmatter。核心不内建提取逻辑（K1），本命令由外部 Agent 执行。
先读规则
- 读 knowledge/ROUTING.md 与 knowledge/applications/INDEX.md（现状路由）
- 读 knowledge/KNOWLEDGE-RULES.md（candidate 治理规则）
- 扫描范围仅限当前需求涉及的仓库路径（最小读取边界）
执行
- 在扫描范围内识别锚点候选：MQ Topic 常量、状态枚举、业务身份常量、对外接口名、核心表名；
- 为每个锚点给出：锚点类型（TOPIC/STATE/BIZ_IDENTITY/INTERFACE/MODEL）、字面值、候选应用、证据文件路径、建议挂载的正式页（applications//…）；
- 只产出候选，不修改任何正式 knowledge/ 页与 ROUTING.md；
- 证据不足者标 confidence: low，禁止推断成事实。
产物
knowledge/candidate/anchors-<YYYY-MM-DD>.md，frontmatter 含：
正文为锚点表：| 类型 | 字面值 | 候选应用 | 建议挂载页 | 证据 | confidence |type: candidate sourceType: ai-assisted engine: rd-extract-routing status: CANDIDATE confidence: medium evidence: - code: <证据文件路径>
何时停止
- 扫描范围无法从需求/ROUTE 推断 → 停止，等人工划定范围。
- 敏感路径（密钥/客户数据/线上配置）一律不读不写。
人工确认
是 — 草稿需 owner review：确认的锚点由人（或后续命令）写入正式页 frontmatter anchors；未确认者留在 candidate，不得被 os wiki routing 合并（合并器只读正式区）。
与索引合并器的关系
- 本适配器产“候选锚点”（外部智能）；
os wiki routing只合并正式区已声明 anchors（确定性核心）；- 二者以 owner review 为界，错误知识不入 ROUTING。
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.
- yesterday First seen · 46 lines · 0 tokens per session scan A f607c0e79a4d
extract-routing is a command published in the GitHub repository alisunstar/OpenSunstar (40 stars, last pushed 18d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 608 tokens. 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 commands, from other repositories
ask
Ask a natural-language question about the repository and get an evidence-backed answer.
ingest-pr
Command "ingest-pr" from Avos-Lab/git-aware-coding-agent, covering avos ingest-pr, usage, description, arguments and options.
history
Get a chronological history of a subject or topic in the repository.
hook-install
Install a git hook for automatic commit sync on push. When installed, every git push automatically syncs pushed commits to Avos Memory, enabling real-time team synchronization.
avos-ask
Search repository memory for engineering context.
avos-history
Get chronological history of a subject in the repository.