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.
git clone --depth 1 https://github.com/zhoubengao/iLinkWrote 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/commands/zhoubengao/ilink/ilink-approve)<a href="https://agentmods.dev/commands/zhoubengao/ilink/ilink-approve"><img src="https://agentmods.dev/badge/commands/zhoubengao/ilink/ilink-approve/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/commands/zhoubengao/ilink/ilink-approve"><img src="https://agentmods.dev/badge/commands/zhoubengao/ilink/ilink-approve.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00000 | $0.01521 |
| Opus 5 | $0.00000 | $0.00760 |
| Sonnet 5 | $0.00000 | $0.00304 |
| Haiku 4.5 | $0.00000 | $0.00152 |
Grade A, and why
ilink-approve 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 10d 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 — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
你现在执行 iLink 的 审核通过(Human-Gate 推进 + Coach 协作复盘) 操作。
本命令自 v1.6.0 起为 Slash Command(不再是 shell 脚本),内部串行执行:校验前置 → 调用 Coach 子流程 → 写入 feedback.md → 推进 Status。Coach 子流程详见
iLink/iLink-root-spec.md§4.7 与iLink/souls/coach.soul.md。
执行任务
在 Story $ARGUMENTS 中找到处于 STAGING 状态的文档,先执行 Coach 子流程,再将其推进到下一阶段。
步骤 1:确认 Story 存在
检查 iLink-doc/$ARGUMENTS/ 目录是否存在,不存在则提示用户先执行 /ilink-init $ARGUMENTS,结束。
步骤 2:按优先级查找 STAGING 文档
依次读取以下文件末尾的 Metadata Status,确定本次 approve 的目标文档:
| 优先级 | 文档 | STAGING 时的目标状态 | 后续命令 |
|---|---|---|---|
| 1 | $ARGUMENTS-review.master.md |
不能自动推进(上游根因) | 告知用户查看 [UPSTREAM_BLOCKERS],与 Designer 讨论修改设计或修改需求定义后结束 |
| 2 | $ARGUMENTS-design.master.md |
PENDING_CODER |
/ilink-coder $ARGUMENTS |
| 3 | $ARGUMENTS-pm.master.md |
PENDING_DESIGNER |
/ilink-design $ARGUMENTS |
如果没有找到 STAGING 文档,告知用户当前没有需要审核的文档,建议执行 /ilink-status $ARGUMENTS 查看详细状态后结束。
注意:QA review 文档处于 STAGING(上游根因)时 SHALL NOT 推进 Status,但 Coach 子流程仍 MUST 执行(评估的是这一轮人类与 AI 的协作,与 Status 是否推进无关)。
步骤 3:调用 Coach 子流程
读取 iLink/souls/coach.soul.md(Coach 角色规范)。
3.1 摘录对话 bracket
按以下规则在当前 Host CLI 主对话中识别 bracket:
- 外层窗口起点:本 Story 的
/ilink-pm $ARGUMENTS命令完成之后的第一个人类 turn - 外层窗口终点:本次
/ilink-approve $ARGUMENTS调用之前的最后一个人类 turn - 内部分界:外层内的
/ilink-design $ARGUMENTSslash 调用 turn
逐 turn 摘录原文,标注格式:
[turn-1] (user)
<原文>
[turn-2] (assistant)
<原文>
...
[--- /ilink-design 分界 ---]
[turn-N] (user)
...
SHALL NOT 改写、概括或选择性剔除任何 turn 的内容。
3.2 计算直接编辑 diff
- 列出
iLink-doc/$ARGUMENTS/.snapshots/目录下所有design.master.*.md文件 - 选择最新的快照(按文件名时间戳降序取第一个)
- 使用
diff -u <最新快照> iLink-doc/$ARGUMENTS/$ARGUMENTS-design.master.md生成 unified diff - 若
.snapshots/目录不存在或为空,记录"无法判定直接编辑(缺少 design 快照)"
若本次目标是 PM 文档(步骤 2 命中优先级 3),跳过 3.2,直接编辑段记为"本段无直接编辑(PM 文档不做编辑检测)"。
3.3 调用 subagent
通过 Agent 工具(general-purpose 类型)调用 subagent,传入以下 bundle(SHALL NOT 传入 master doc、requirement.md、源码):
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.
- 10d ago First seen · 131 lines · 0 tokens per session scan A f89bdbd6950e
ilink-approve is a command published in the GitHub repository zhoubengao/iLink (22 stars, last pushed 3mo ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,521 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
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.