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 FirenzeClaw/kimi-session-orchestrator --skill loop-orchestratorgit clone --depth 1 https://github.com/FirenzeClaw/kimi-session-orchestratorWrote 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/firenzeclaw/kimi-session-orchestrator/loop-orchestrator)<a href="https://agentmods.dev/skills/firenzeclaw/kimi-session-orchestrator/loop-orchestrator"><img src="https://agentmods.dev/badge/skills/firenzeclaw/kimi-session-orchestrator/loop-orchestrator.svg" alt="Measured on agentmods" 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 | $0.00076 | $0.01535 |
| Opus 5 | $0.00038 | $0.00767 |
| Sonnet 5 | $0.00015 | $0.00307 |
| Haiku 4.5 | $0.00008 | $0.00153 |
Grade A, and why
loop-orchestrator 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 7d 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 — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Loop Orchestrator — PM 自主循环编排
⛔ 加载即执行——启动协议
此 skill 加载后，必须完成以下步骤再处理任何用户请求：
Auto 检测
| 模式 | 检测方式 | Q1-Q4 交互方式 |
|---|---|---|
| Auto | 系统提示含 Auto permission mode is active |
纯文本提问，提示用户 /auto 可退出获得交互式选项 |
| 非 Auto | 无上述提示 | 使用 AskUserQuestion 工具，每次一个问题 |
第一轮：Q1 — 目标采集
若用户调用时已带目标描述（如 /loop-orchestrator 审查 demo/ 全部模块...）→ 跳过 Q1，直接进入 Q2。
否则 → 纯文本问询："最终目标是什么？（可含路径、模块、验收标准）"
第二轮：Q2 — 模式选择
- A: 实施循环（Implement） → Read
~/.kimi-code/skills/loop-orchestrator/guide-loop-implement.md - B: 验收循环（Verify） → Read
~/.kimi-code/skills/loop-orchestrator/guide-loop-verify.md - C: 混合（Hybrid） → 先验收现状 → 再实施缺失。先 Read
~/.kimi-code/skills/loop-orchestrator/guide-loop-verify.md→ 验收完成后 Read~/.kimi-code/skills/loop-orchestrator/guide-loop-implement.md
第三轮：Q3 — 并行策略
- A: 单 session 串行 → Read
~/.kimi-code/skills/loop-orchestrator/guide-loop-core.md - B: 多 session 并行 → Read
~/.kimi-code/skills/loop-orchestrator/guide-loop-parallel.md
第四轮：Q4 — 验收标准
"验收标准是什么？按可验证条件逐条列出。"
- 若用户未给出 → PM 自行从项目 spec/AGENTS.md 提取 + 展示确认
- 若用户给出 → PM 逐条确认可验证性
每条标准必须独立可验。示例："src/types.ts 中所有类型与 spec §3.2 字段定义一致" ✅ / "代码没问题" ❌
第五轮：确认概要 → 进入 PM 自主编排
输出确认卡片：
目标: <摘要>
模式: 实施/验收/混合
并行: 单/多 session
验收标准: <N 条>
用户确认后进入自主编排。此后 PM 自主全权决策，里程碑自动汇报，不降级目标。
⛔ PM 硬边界
| ✅ 允许 | ❌ 禁止 |
|---|---|
create_session / execute_prompt |
Edit / Write（绝不碰文件） |
poll_session / list_io_records / read_session_log |
Bash（文件操作/构建/测试/代码执行） |
grade_step / memory_* |
自行降级目标（绝对目标铁律） |
approve_tool / deny_tool |
|
| Bash（仅限：后台轮询 poll_command / 读日志 / 读 output_path） | |
| ⛔ 每次 execute_prompt 后必须立即 Bash(run_in_background=true) 执行 poll_command | 跳过此步骤 = 流程违规，task session 回复永远丢失 |
核心铁律
⛔ execute_prompt → 立即 Bash(run_in_background=true) → 确认 task_id。三步不可拆分、跳过、延迟。
| 规则 | 违反后果 |
|---|---|
| execute_prompt 后立即 Bash 后台轮询（用 poll_command 原文） | 跳过 → task session 回复永远丢失 |
| 即发即返，不阻塞 | MCP 超时截断 |
| 后台 Bash 轮询 | 零 token 等待 |
| ⛔ 逐条注入（一次一个操作指令） | 多操作合一 → 注意力稀释、PM 无法定位错误 |
| 跨模块必须分 session | 上下文污染 |
| session 复用优先 | 新建仅限：context_tokens > 36K / 质量下降 / 模块切换 |
What ships with it
8 files 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.
- 7d ago First seen · 118 lines · 76 tokens per session scan A f4a82589cf4f
loop-orchestrator is a skill published in the GitHub repository FirenzeClaw/kimi-session-orchestrator (0 stars, last pushed 12d ago), licensed MIT. It adds 76 tokens to every session and 1,535 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 2026-08-31.
Other skills, from other repositories
workflow
Creates durable, resumable workflows using Vercel's Workflow SDK. Use when building workflows that need to survive restarts, pause for external events, retry on failure, or coordinate multi-step operations over time. Triggers on mentions of "workflow", "durable functions", "resumable", "workflow sdk", "queue"…
migrating-workflow-v4-to-v5
Upgrades an app from Workflow SDK 4.x to 5.0. Use when bumping the workflow / @workflow/ dependencies to v5, or when hitting removed v4 APIs — runStep, stepEntrypoint, workflow/internal/private, @workflow/core/private, writeToStream / closeStream / readFromStream on a World, world.steps.get without a runId…
migrating-world-v4-to-v5
Upgrades a custom Workflow SDK World implementation from the v4 spec to v5. Use when a package implements the World interface from @workflow/world and is moving to 5.x — event IDs that are ULIDs rather than slot positions, Event id is not slot-numbered at replay time, a specVersion the runtime refuses, writeToStream /…
migrating-to-workflow-sdk
Migrates Temporal, Inngest, Trigger.dev, and AWS Step Functions workflows to the Workflow SDK. Use when porting Activities, Workers, Signals, step.run(), step.waitForEvent(), Trigger.dev tasks / wait.forToken / triggerAndWait, ASL JSON state machines, Task/Choice/Wait/Parallel states, task tokens, or child workflows.
streamsql
A Chinese-language guide to StreamSQL, a SQL-based way to process continuously arriving data in RuleGo. It covers filtering, transformations, time windows, joins, change tracking, cumulative values, and event-pattern matching.
workflow-init
Install and configure Vercel Workflow SDK before it exists in nodemodules. Use when the user asks to "install workflow", "set up workflow", "add durable workflows", "configure workflow sdk", or "init workflow" for Next.js, Express, Hono, Fastify, NestJS, Nitro, Nuxt, Astro, SvelteKit, or Vite.