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 noho/code-is-cheap --skill sub-agentsgit clone --depth 1 https://github.com/noho/code-is-cheapWrote 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/noho/code-is-cheap/sub-agents)<a href="https://agentmods.dev/skills/noho/code-is-cheap/sub-agents"><img src="https://agentmods.dev/badge/skills/noho/code-is-cheap/sub-agents/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/noho/code-is-cheap/sub-agents"><img src="https://agentmods.dev/badge/skills/noho/code-is-cheap/sub-agents.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 | $0.00058 | $0.02697 |
| Opus 5.5 | $0.00023 | $0.01079 |
| Sonnet 5.5 | $0.00012 | $0.00539 |
| Haiku 4.5 | $0.00006 | $0.00270 |
Grade A, and why
sub-agents 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 6d 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 — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Sub Agents
通过 runner 子进程调用外部 Claude Code / Codex 子 Agent。普通任务直接派发，等待实际退出，读取汇总里的 最终答复，并按任务需要使用或核对成果。项目明确要求的审核、取证和工作流规则仍然适用。
直接调用
| Runtime | Runner | Providers |
|---|---|---|
| Claude Code | claude-agent-run |
ds-flash mimo mimo-fast mimo-flash qwen kimi glm glm-flash local hy |
| Codex | codex-agent-run |
ds-flash mimo mimo-fast mimo-flash qwen kimi glm glm-flash local gpt-6-astra gpt-6-astra-x gpt-6-astra-max gpt-6-sol gpt-6-sol-x gpt-6-sol-max gpt-6-luna business |
Astra 与 Sol 均提供默认 high、-x xhigh 和 -max max 三档；Sol 上游为 gpt-6.1-sol。这些 provider 使用订阅登录，不需要 API key 或 shim；business 同步 Sol 默认 high，Luna 保持 max。旧 gpt-6-astra-xhigh 不再是受支持的 runner provider。
每次显式传 --cwd 的 workspace 绝对路径，避免依赖总控当前目录。接口不熟悉时查看所选 runner 的 --help / --list-providers。
codex-agent-run --provider mimo --cwd /absolute/workspace --prompt-file /absolute/task.md
claude-agent-run --provider ds-flash --cwd /absolute/workspace --prompt "检查 src/parser.py 对空输入的处理，只报告发现，不修改文件。"
runner 自动检查所选 provider、workspace、任务输入、launcher 及输出路径；非 Git workspace 的 Codex 自动追加
--skip-git-repo-check。默认一次性运行，不持久化 session。默认自动分配私有日志目录和 Claude instance；
调用方无需提供 label、instance、日志路径或单独运行预检。需要指定日志位置时使用 --output / --stderr /
Codex --last-message，这些必须是新文件；显式 - 仅用于 --detail，汇总模式明确拒绝。
独立上下文交接
--prompt / --prompt-file 的任务在新的独立 Agent 上下文执行，不继承总控的对话、裁决或用户授权。
相同 cwd 或 provider 也不传递上下文；runtime 自身可能加载 workspace 的 AGENTS.md 等本地指令和配置。
用自然语言提供这项任务需要的背景、已决事项、授权、输入/版本、依赖、可写范围和交付要求，或给出可定位的文件路径。 复杂任务明确范围和完成信号；简单任务无需固定标题或填写 N/A。不要使用只有父上下文知道含义的“刚才”“照旧”。 不要求子 Agent 自报 runtime/provider/model。只交接必要信息，不转发无关的历史或敏感信息。 缺少必要输入、授权或工具时应报告具体缺项，不能猜测或自动扩大权限。
要求交付文件时，把目的及文件路径告诉子 Agent，并要求在最终答复给出产物路径和未完成事项。 修改既有源码无需另写一份报告描述修改；调用方可直接查看源码和任务要求的验证。
托管执行与等待
派发从总控沙箱外运行，避免父级沙箱妨碍 runtime 初始化或子 Agent 的工具：
- Codex 总控：每个任务用一次独立的
exec_command，设置sandbox_permissions: "require_escalated"并说明任务。 返回 session_id 时，用对应write_stdin收集，直到取得退出码。 - Claude 总控：以独立裸 runner 命令调用 Bash，使其命中
sandbox.excludedCommands；准备工作在之前完成。 后台运行用run_in_background: true，用托管任务句柄/完成通知收集。不要用 shell&、管道或复合命令派发。
What ships with it
2 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.
- 6d ago Changed · +2 lines 39c928edc989
- 9d ago Changed 2dc94bf4a391
- 11d ago Changed · -8 lines a40660709e3c
- 16d ago Changed · +11 lines cb812b367be3
- 19d ago Changed · -211 lines · +11 tokens per session e73839dbb693
- 20d ago Changed · +7 lines 6c92485cb41b
- 21d ago Changed · +27 lines 417ebb20c657
- 29d ago Changed · +66 lines ed2e7ba2ecab
- 1mo ago Changed a60ae4ac65ba
- 1mo ago First seen · 233 lines · 47 tokens per session scan A 2c3d1118772b
sub-agents is a skill published in the GitHub repository noho/code-is-cheap (94 stars, last pushed 7d ago), licensed Apache-2.0. It adds 58 tokens to every session and 2,697 once invoked, about $0.0002 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 2026-09-19.
Other skills, from other repositories
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…