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/double-coding-lab/flow2spec/f2s-kb-addnpx skills add double-coding-lab/Flow2Spec --skill f2s-kb-addgit clone --depth 1 https://github.com/double-coding-lab/Flow2SpecWhat 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 | $0.00055 | $0.02691 |
| Opus 5 | $0.00028 | $0.01345 |
| Sonnet 5 | $0.00011 | $0.00538 |
| Haiku 4.5 | $0.00006 | $0.00269 |
Grade A, and why
f2s-kb-add 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 2d 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 — 133 lines — stays where its author put it; the contents beside it link to each section on GitHub.
执行口径：本技能只维护
.Knowledge，不改配置根rules/skills。
编排（主 / 子 agent）
subAgent/switchAgentVerification两字段语义以统一入口为唯一事实源：Cursor/Claude 读配置根rules/f2s-flow2spec-unified-entry.*；Codex 读.codex/topics/f2s-flow2spec-unified-entry.md（与上同源，flow2spec init镜像）。- 默认不拆子：主会话全流程完成；低于阈值时拆子收益低于 context 切换成本。
- 拆子阈值（仅当
subAgent=true且任一满足）：① 输入路径 ≥ 5；② 单源文件 > ~3000 行；③ 多路径总量 > ~10000 行。 - 拆子策略（仅在达到拆子阈值且
subAgent=true时启用）：- B 模式（默认，单轮并行）：主先产出「inventory（待解析源文档路径清单 + 核心能力名，主手写，禁止子 agent 自行增删）」+「扫描契约（每个源读哪些章节 / 行号范围、禁扫目录、统一产出字段与表头）」→ 子 agent 并行只读按表填写 → 主一轮合并 + 去重 → 写
.Knowledge/stock-docs/<方案名>_初稿.md→ 主做用户确认与验收。适合源边界较清晰、中等规模、希望尽快出一版。 - C 模式（大仓 / 高风险，多轮纠偏）：在 B 之前或替代 B 首轮 —— 主先做 inventory → 子并行交表 → 主专做一轮对表（标重合 / 矛盾 / 缺依赖 / 跨源边界）→ 必要时对矛盾点补派小任务或主自读关键点 → 最后主写 / 改定稿。适合多 workspace / monorepo、目录极深、源路径 > 20 条、首轮子表矛盾或空洞明显、多源叙述重合或矛盾严重的场景。
- 切换判据（任一成立即切到 C）：多 workspace / monorepo；目录极深或源路径 > 20 条；首轮子表矛盾 / 空洞明显；多源叙述重合 / 矛盾严重。
- B 模式（默认，单轮并行）：主先产出「inventory（待解析源文档路径清单 + 核心能力名，主手写，禁止子 agent 自行增删）」+「扫描契约（每个源读哪些章节 / 行号范围、禁扫目录、统一产出字段与表头）」→ 子 agent 并行只读按表填写 → 主一轮合并 + 去重 → 写
- 子交付硬约束：子 agent 不得自行裁剪源路径范围，必须按主手写 inventory 执行；交付按「子交付 YAML schema」（字段：
source/scope/capabilities/cross_refs/pending），禁止散文式回传；子不得写manifest-routing.json/.Knowledge/index.md；子不得单独宣布「已进知识库」。 - 主必控：重合判定、终稿定稿、
f2s-kb-build调度、整体验收。 - 写权硬约束：
manifest-routing.json与.Knowledge/index.md恒由主 agent 落盘。 - 落盘侧自验。
f2s-kb-add：多文件聚合 -> 初稿 -> 终稿 -> 知识路由同步
使用时机
- 某能力已在代码中落地，但信息分散在多个文件，需沉淀为可检索知识。
- 与
f2s-doc-arch区分：doc-arch产出架构初稿；doc-add产出“已落地能力”知识沉淀链路。
输入
| 参数 | 必填 | 说明 |
|---|---|---|
| 文件路径列表 | 是 | 一个或多个路径（空格/换行/@）；支持源码、配置、文档 |
| 方案名 | 否 | 用于生成 <方案名>_初稿.md、<方案名>_终稿.md |
| 初稿/终稿路径 | 否 | 默认放 .Knowledge/stock-docs/ |
无有效路径时中止并要求用户补充。
步骤 0：重合判定（重要）
执行前先对照：
.Knowledge/index.md.Knowledge/topics/*.md.Knowledge/stock-docs/*.md
若已有同主题沉淀，优先原位更新，避免重复主题和重复索引行。
步骤 0.5：多模块检测（输入路径 ≥ 2 时必须执行）
- 目录聚合：按路径中的功能层目录（如
src/<模块名>/、顶层目录名）对文件分组。 - 判定规则（满足任一即判定为「多模块」）：
- 文件分属 ≥ 2 个不同顶层功能目录（如
auth/、payment/）； - 用户在输入中明确提及「多个功能 / 不同模块 / 分别处理」等；
- 文件名前缀明显不同且无共同父目录。
- 文件分属 ≥ 2 个不同顶层功能目录（如
- 单模块（未触发判定）：不中断，继续步骤 1，按现有单输出逻辑生成
<方案名>_初稿.md。 - 多模块（触发判定）：暂停，向用户展示分组结果，并询问：
- 方案 A（推荐）：按模块分别生成知识文件 → 每组独立走步骤 1→2→3→4，各自产出
<模块名>_初稿.md/<模块名>_终稿.md； - 方案 B（合并）：忽略模块边界，合并生成一份
<方案名>_初稿.md（原有行为）。 - 禁止在未获用户明确选择前默认走方案 B 继续执行。
- 方案 A（推荐）：按模块分别生成知识文件 → 每组独立走步骤 1→2→3→4，各自产出
- 单模块但 stock-doc 体量大：若单份输入文档或聚合后的源码超过 300–500 行，或涵盖 3 个以上不相干职责域，建议向用户提示"可拆成多份 focused stock-doc，各自对应独立 topic"；用户确认继续则不阻断，但在输出摘要中记录"建议后续拆分"。
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.
- 2d ago First seen · 133 lines · 55 tokens per session scan A b03d7051cbf0
f2s-kb-add is a skill published in the GitHub repository double-coding-lab/Flow2Spec (47 stars, last pushed 5d ago), licensed MIT. It adds 55 tokens to every session and 2,691 once invoked, about $0.0003 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.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
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…