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 skills/yunzeforbetter/castflow/code-pipeline-skillnpx skills add yunzeforbetter/CastFlow --skill code-pipeline-skillgit clone --depth 1 https://github.com/yunzeforbetter/CastFlowWrote 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/yunzeforbetter/castflow/code-pipeline-skill)<a href="https://agentmods.dev/skills/yunzeforbetter/castflow/code-pipeline-skill"><img src="https://agentmods.dev/badge/skills/yunzeforbetter/castflow/code-pipeline-skill.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 | $0.00049 | $0.01965 |
| Opus 5 | $0.00024 | $0.00983 |
| Sonnet 5 | $0.00010 | $0.00393 |
| Haiku 4.5 | $0.00005 | $0.00197 |
Grade A, and why
code-pipeline-skill 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 — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Code Pipeline 工作流
code-pipeline-skill 是一个复合组件：它组合 shared agents、模块配对执行单元与可选 skill，在 pipeline 执行期间维护自己的协议与运行态投影；它不定义 shared components 的本体规则。
场景入口
- 标准模式：单模块、双模块、依赖链短，但仍需要 Step 1 / 3 / 4 / 5 的最小闭环
- 复杂系统模式：3+ 模块、共享契约重、需要 wave / dispatch / checkpoint / sub-pipeline
- 简单局部改动：若一个 skill 或单个 agent 就能收敛，不值得启 pipeline
- 但只要输入包含 PRD / 外部需求文档 / 设计稿，或执行者尚未掌握该功能域的现有承载面，就不得以“简单局部改动”为由跳过 pipeline；至少先完成一次 Step 1
AI 读取顺序
SKILL_MEMORY.mdconfig/step_contracts.mdconfig/pipeline_protocol.md- 进入
L1+、多模块协作或需要 Freeze / Closure / Coverage 时，再读config/handoff_protocol.md - 需要 runtime state 真身时，读
config/runtime_state_schema.md - 进入复杂系统模式后，再按需读取对应主定义页：
- 总模型：
architecture/orchestration-model.md - 状态：
architecture/artifact-state-machine.md - barrier / wave / dispatch：
architecture/barrier-and-wave-scheduling.md - 子 pipeline：
architecture/subpipeline-strategy.md - stalled / recovery：
architecture/stall-recovery.md - 局部 / 全局验证：
architecture/verification-architecture.md
- 总模型：
- 需要模板或判例时读取
EXAMPLES.md与examples/*
Step 1 快速门
- 在
code-pipeline模式下，Step 1 的正文必须显式出现独立的### CapabilityScanblock，不能用 Phase 叙事、PRD 摘要或一句“已检索”代替 CapabilityScan至少要写出MatchedCapabilities、CandidateHosts、Evidence、Recommendation；即使未命中可复用承载，也必须显式写None/未命中，不能省略Evidence只接受项目代码路径、符号、检索范围与命中说明；PRD、用户口述、设计说明只能定义scan scope，不能单独充当源码证据- 未满足以上条件时，主流程只能维持
Discovering或PendingDecision，不得宣布 Step 1 完成，更不得进入 Step 2 / Step 3
工作流总览
Step 1-9 的唯一 Step 收口页与 Step 级强规则入口见 config/step_contracts.md；执行期的精确调度合同、状态迁移与读写路径见 config/pipeline_protocol.md。本页只回答三件事：什么时候启 pipeline、每一步由谁执行、接下来该深读哪里。
Step 1 内部允许按 DecompositionSnapshot（拆解快照） -> CapabilityScan（能力探寻） -> ArtifactBinding（产物绑定） -> DecisionSynthesis（决策综合）做认知分治；scan 类任务可以拆给只读 subagent，但 decision_status、Handoff Level Decision、Freeze Recommendation 仍只由 Step 1 聚合层写回。
Step 导航
| Step | 执行单元 | 导航信息 | 深读 |
|---|---|---|---|
| Step 1 | requirement-analysis-agent |
先产出 DecompositionSnapshot；复杂认知优先拆给只读 scan subagent；若 CapabilityScan 无独立 block 或缺少源码 Evidence，视为 Step 1 未完成；若存在路线分歧，必须停在决策态，不能直接推进实现 |
config/step_contracts.md、config/pipeline_protocol.md |
| Step 2 | requirement-analysis-agent |
只在路线已收敛且命中冻结条件时进入 | config/step_contracts.md、config/pipeline_protocol.md、config/handoff_protocol.md |
| Step 3 | 模块配对执行单元 / main agent（L0） |
负责实现；进入前必须确认决策态已解决且 Freeze Gate 已过 | config/step_contracts.md、config/pipeline_protocol.md |
| Step 4 | integration-matching-agent |
只做依赖闭合验证，不改代码 | config/step_contracts.md、config/handoff_protocol.md |
| Step 5 | pipeline-verify-agent |
只做 coverage / verdict 决策，不改代码 | config/step_contracts.md、config/pipeline_protocol.md、config/handoff_protocol.md |
| Step 6 | 模块配对执行单元 / main agent |
只补 CompletableBlocks，补完后至少回 Step 4 |
config/step_contracts.md、config/handoff_protocol.md |
| Step 7 | debug-skill |
只在需要边界条件验证时进入 | config/step_contracts.md |
| Step 8 | profiler-skill |
只在需要性能诊断时进入 | config/step_contracts.md |
| Step 9 | main agent |
必做收尾；未完成 run_id / signal 清理不得结束 pipeline | config/step_contracts.md、config/pipeline_protocol.md |
What ships with it
22 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.
- architecture/artifact-state-machine.md 2.3 KB
- architecture/barrier-and-wave-scheduling.md 1.9 KB
- architecture/orchestration-model.md 1.7 KB
- architecture/README.md 1.3 KB
- architecture/stall-recovery.md 1.7 KB
- architecture/subpipeline-strategy.md 1.4 KB
- architecture/verification-architecture.md 2.0 KB
- config/defaults.json 105 B
- config/handoff_protocol.md 5.4 KB
- config/params.schema.json 794 B
- config/pipeline_protocol.md 25 KB
- config/runtime_state_schema.md 3.2 KB
- config/step_contracts.md 12 KB
- EXAMPLES.md 28 KB
- examples/complex-system-entry.md 1.5 KB
- examples/stall-recovery-example.md 1.9 KB
- examples/subpipeline-example.md 1.4 KB
- examples/wave-planning-example.md 2.6 KB
- ITERATION_GUIDE.md 2.8 KB
- README.md 3.8 KB
- scripts/pipeline_merge.py 8.8 KB runs code
- SKILL_MEMORY.md 4.6 KB
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 First seen · 101 lines · 49 tokens per session scan A e6be09a629a2
code-pipeline-skill is a skill published in the GitHub repository yunzeforbetter/CastFlow (108 stars, last pushed 2mo ago), licensed MIT. It adds 49 tokens to every session and 1,965 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
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…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…
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…
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…